![ReviewArticle - Hindawi Publishing Corporationdownloads.hindawi.com/journals/bmri/2011/527201.pdf2 Journal ofBiomedicineand Biotechnology locations,thereareclearpersistingfunctionaldeficits[8–10].](https://static.fdocuments.in/doc/165x107/5e35e06aa654b36d62499984/reviewarticle-hindawi-publishing-2-journal-ofbiomedicineand-biotechnology-locationsthereareclearpersistingfunctionaldeicits8a10.jpg)
ReviewArticle - Hindawi · 2019. 11. 7. · Geological strength index From the description of...
18
Review Article Review of the Geological Strength Index (GSI) as an Empirical Classification and Rock Mass Property Estimation Tool: Origination, Modifications, Applications, and Limitations Sajjad Hussian , 1 Noor Mohammad, 1 Zahid Ur Rehman, 1 Naseer Muhammad Khan, 2,3 Khan Shahzada, 4 Sarfraz Ali, 5 Muhammad Tahir, 1 Salim Raza, 1 and Saira Sherin 1 1 Department of Mining Engineering, University of Engineering and Technology, Peshawar, Pakistan 2 Department of Mining Engineering, Balochistan University of Information Technology and Management Sciences (BUITMS), Balochistan, Pakistan 3 School of Mines, China University of Mining and Technology, Xuzhou, China 4 Department of Civil Engineering, University of Engineering and Technology, Peshawar, Pakistan 5 Advanced School of Geo-Mechanical Engineering (SAGE), National University of Science and Technology, Islamabad, Pakistan Correspondence should be addressed to Sajjad Hussian; [email protected] Received 7 November 2019; Revised 19 July 2020; Accepted 30 July 2020; Published 24 August 2020 Academic Editor: Paolo Castaldo Copyright©2020SajjadHussianetal.isisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. e geological strength index (GSI) is one of the most exceptional rock mass classification system which is used to evaluate very weak and highly jointed rock mass by different approaches and related to rock mass geomechanical properties including generalized Hoek & Brown constants, deformation modulus, strength properties, and Poisson’s ratio for an appropriate design of tunnels, caverns, and other engineering structures. e distinctiveness of this system over the rock mass rating (RMR), Q-system, and other empirical methods is as follows: it utilized field observations, blockiness of rock mass, and surface joint characteristics during the evaluation process of rock mass and efficiently espoused as an empirical tool for estimation of geomechanical properties of rock mass required for pre-post stability of engineering structures using numerical modeling. is study presents the review of the 19 years of research studies conducted by different researchers about the GSI in a systematic way, i.e., origination, modifications, applications, and limitations. Furthermore, this study will provide a better understanding to field professionals (geologists, mining and civil engineers) about the qualitative and quantitative estimation of the GSI and its application as an empirical estimating tool for an appropriate design of engineering structures in rock mass environments. 1.Introduction e complex nature and uncertain behavior of rock mass due to anisotropic and presence of unconformities make the rock mass as difficult material for empirical and numerical modeling [1–5]. Modeling of such complex nature and un- certain behavior of rock mass in a precise way to decrease the uncertainty associated with the characterization process is quite complicated job [6]. Hudson and Feng [4] presented four primary modeling methods (A, B, C, and D) and eight submethods arranged in two levels for the solution/modeling of complex rock mass conditions using forward and back analysis-based design as shown in Figure 1. e design process of engineering structure is started from site investigations and followed by the main four methods (A, B, C, and D) for completion of the project. At the preliminary stage of any rock engineering project, the detail data about geology and geotechnics of the area are not available for the proper design of engineering structures. erefore, to deal with this situation, different rock mass classification systems called empirical methods of design have been developed for completion of the required design at the preliminary stage of the project [7–10]. e empirical methods of design are considered very helpful in rock mass characterization and classification into different classes having similar characteristics for quickly understanding and Hindawi Advances in Civil Engineering Volume 2020, Article ID 6471837, 18 pages https://doi.org/10.1155/2020/6471837
Transcript of ReviewArticle - Hindawi · 2019. 11. 7. · Geological strength index From the description of...
Review ArticleReview of the Geological Strength Index (GSI) as an EmpiricalClassification and Rock Mass Property Estimation ToolOrigination Modifications Applications and Limitations
Sajjad Hussian 1 Noor Mohammad1 Zahid Ur Rehman1 Naseer Muhammad Khan23
Khan Shahzada4 Sarfraz Ali5 Muhammad Tahir1 Salim Raza1 and Saira Sherin1
1Department of Mining Engineering University of Engineering and Technology Peshawar Pakistan2Department of Mining Engineering Balochistan University of Information Technology and Management Sciences (BUITMS)Balochistan Pakistan3School of Mines China University of Mining and Technology Xuzhou China4Department of Civil Engineering University of Engineering and Technology Peshawar Pakistan5Advanced School of Geo-Mechanical Engineering (SAGE) National University of Science and Technology Islamabad Pakistan
Correspondence should be addressed to Sajjad Hussian engrsajjaduetpeshawaredupk
Received 7 November 2019 Revised 19 July 2020 Accepted 30 July 2020 Published 24 August 2020
Academic Editor Paolo Castaldo
Copyright copy 2020 SajjadHussian et al+is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
+e geological strength index (GSI) is one of the most exceptional rock mass classification system which is used to evaluate veryweak and highly jointed rock mass by different approaches and related to rock mass geomechanical properties includinggeneralized Hoek amp Brown constants deformation modulus strength properties and Poissonrsquos ratio for an appropriate design oftunnels caverns and other engineering structures +e distinctiveness of this system over the rock mass rating (RMR) Q-systemand other empirical methods is as follows it utilized field observations blockiness of rock mass and surface joint characteristicsduring the evaluation process of rock mass and efficiently espoused as an empirical tool for estimation of geomechanicalproperties of rock mass required for pre-post stability of engineering structures using numerical modeling+is study presents thereview of the 19 years of research studies conducted by different researchers about the GSI in a systematic way ie originationmodifications applications and limitations Furthermore this study will provide a better understanding to field professionals(geologists mining and civil engineers) about the qualitative and quantitative estimation of the GSI and its application as anempirical estimating tool for an appropriate design of engineering structures in rock mass environments
1 Introduction
+e complex nature and uncertain behavior of rock mass dueto anisotropic and presence of unconformities make the rockmass as difficult material for empirical and numericalmodeling [1ndash5] Modeling of such complex nature and un-certain behavior of rock mass in a precise way to decrease theuncertainty associated with the characterization process isquite complicated job [6] Hudson and Feng [4] presentedfour primary modeling methods (A B C and D) and eightsubmethods arranged in two levels for the solutionmodelingof complex rock mass conditions using forward and backanalysis-based design as shown in Figure 1
+e design process of engineering structure is startedfrom site investigations and followed by the main fourmethods (A B C and D) for completion of the project Atthe preliminary stage of any rock engineering project thedetail data about geology and geotechnics of the area are notavailable for the proper design of engineering structures+erefore to deal with this situation different rock massclassification systems called empirical methods of designhave been developed for completion of the required designat the preliminary stage of the project [7ndash10] +e empiricalmethods of design are considered very helpful in rock masscharacterization and classification into different classeshaving similar characteristics for quickly understanding and
HindawiAdvances in Civil EngineeringVolume 2020 Article ID 6471837 18 pageshttpsdoiorg10115520206471837
the construction of underground engineering structures[11 12] based on discontinuities and physiomechanicalproperties of rock mass for capturing the real image of the insitu conditions at the preliminary stage for effective design ofengineering structures [13 14] Numbers of empiricalmethods of design have been developed based on civil andmining engineering related case studies by various re-searchers in order to ensure the curative of the complexnature of rock mass for the design of tunnels caverns rockslopes and general purposes as shown in Table 1
+ese empirical methods have wide applications in thecivil and mining fields Each empirical approach used de-fined input parameters including field observations andlaboratory testing for solving different aspects of rock en-gineering projects [36 37] In the aforementioned empiricalmethods RMR and Q-system are widely used classificationsystems for various purposes in rock engineering andconsidered as a base for the development of other classifi-cation systems [28 33 38]
+e rock mass rating (RMR) and tunneling quality index(Q-system) used rock quality designation (RQD) as an in-tegral input parameter in the evaluation of rock mass qualityfor the design of a support system for underground struc-tures [39] +e value of RQD is determined by core drillingwhich is a very tedious procedure [17] +e procedure iscost-effective and the value of RQD for a very weak rock isusually considered zero [17] +is condition leads to anestimation of the inappropriate value of RMR and Q-systemfor the classification of rock mass [40] +e RMR and
Q-system classification systems are mostly used for simplerock mass behavior and these are less reliable to be used insqueezing defined structural failures rock bursting undervery high-stress conditions and estimation of rock massproperties for numerical modeling and less helpful ingaining information about the sequential installation ofsupports for controlling of progressive failure in weak rockmass conditions during tunneling construction [41]Keeping in view of the above weaknesses of RMR andQ-system the GSI rock mass classification system was de-veloped After development of the GSI system differentresearchers worldwide have conducted research studiesabout the various aspects of the GSI system to accommodateand evaluate weakest jointed and heterogeneous rock massenvironment for design of engineering structures and nu-merical modeling [11 13 41ndash57] +e research conducted sofar on GSI is not comprehensive reviewed in a systematicway therefore it is very important to carry out a systematicand comprehensive review of the research conducted aboutthe various aspects of GSI system in order to provide a betterunderstanding to field professionals (geologists mining andcivil engineers) about the GSI system and its application asan empirical classification and rock mass property estima-tion tool for an appropriate design of engineering structuresand numerical modeling
In this study a review of 19 years of research conductedby different researchers about the various aspects of the GSIis presented in a systematic way +e review about the GSI isdivided into four categories ie origination modifications
Siteinvest-igation
Use ofpre-existing
standardmethods
Precedent typeanalyses and
modifications
Analyticalmethods
stress-based
Rock massclassificationRMR Q GSI
Basicnumerical
methods FEMBEM DEM
hybrid
Extendednumericalmethods
fully coupledmodels
Level 11 1 mapping
Level 2Not 1 1 mapping
Integratedsystem
approachesinternet-based
Databaseexpert
systems andother systemapproaches
Design based on back analysisDesign based on forward analysis
Construction
Method A Method B Method C Method D
Objective
Figure 1 Summary of different methods for the solution of rock engineering problems [4]
2 Advances in Civil Engineering
applications and limitations in order to provide a betterunderstanding to field professionals (geologists mining andcivil engineers) about the qualitative and quantitative esti-mation of the GSI for assessment of rock mass and use theGSI as an empirical estimating tool for an appropriate designof engineering structures in rock mass environments
2 Origination of the Geological StrengthIndex (GSI)
After reviewing the empirical methods of design as pre-sented in Table 1 the following grounds are identified for thedevelopment of the GSI
(a) RQD is an integral part of RMR and Q-system Inweak rock mass conditions RQD value is taken aszero using Deree et al procedure and equation (1)[58] +is condition leads to an estimation of theinappropriate value of RMR and Q-system for rockmass classification [40]
(b) RMR Q-system and other rock mass classificationsystems are inconvenient to provide information aboutthe requirement of the support system in complextunneling conditions for controlling progressive failureand design of proper sequentially installed temporarysupports and on the contrary the numerical methodsare considered very effective to give detail informationabout this situation +e numerical methods used thephysiomechanical and deformation properties of rockmass surrounding the tunnel as input parameters
+ese properties of the rock mass are difficult tomeasure in the field due to uncertain rock massconditions and challenging procedure [7 40 58ndash60]Researchers are now focused on estimating the rockmass properties rather than finding in the field [61] Inthis regard the first attempt was made to estimate therock mass properties by Hoek and Brown in 1980 [61]Hoek and Brown developed failure criteria to estimatethe rock mass properties It was realized after thedevelopment of Hoek and Brownrsquos failure criteria thatthis failure criterion should be linked with actualgeological observations of the field to gain its practicalvalue [41]+us in this context the development of thegeological strength index (GSI) was started in Torontowith taking engineering geology as input from DavidWood in 1992 by Hoek et al to classify weak andjointed rock mass environments
RQD 1113944(length of core piecesgt 10 cm length)
times100
total length of core run
(1)
3 Geological Strength Index (GSI)
+e geological strength index (GSI) is a rock mass char-acterization tool developed for the design of tunnelscaverns and other underground structures based on fieldobservations including geological data about rock massinputs from qualified and expertise field geologists
Table 1 Empirical methods for rock mass classification and design of engineering structures in rock mass
S no Empirical method Year Applications Authors Reference1 Rock load 1946 Tunnels with steel support Terzaghi [15]2 Stand-up time 1958 Tunneling Lauffer [16]3 Rock quality designation (RQD) 1963 1966 Tunneling Deere and Miller [17 18]4 Rock structure rating (RSR) 1972 Tunneling Wickham et al [19]
5 Rock mass rating (RMR) 1973 (modifications1989) Tunnels mines slopes Bieniawski [20 21]
6 Tunneling quality index (Q) 1974 (lastmodification 2002)
Tunnels minesfoundations Barton et al [22]
7 New Austrian tunneling method (NATM) 1974 Tunneling Pacher andRabcewicz [23]
8 Size strength classification 1975 Tunneling Franklin [24]9 Basic geotechnical classification (BGC) 1981 General ISRM [25]10 Rock mass strength (RMS) 1982 Metal mining Stille and Groth [26]
11 Unified rock mass classification system(URCS) 1984 General Williamson and
Kuhn [27]
12 Rock mass index (RMi) 1996 Tunneling Palmstrom [28]
13 Geological strength index (GSI) 1997 All undergroundexcavations Hoek and Kaiser [29]
14 Rock tunneling quality index by TBMexcavation (QTBM)
1999 TBM tunnels Barton [30]
15 Continuous rock mass rating (CRMR) 2003 General Sen and Bahaaeldin [31]
16 Rock mass excitability (RME) 2006 TBM tunnels Bieniawski vonPrein et al [32]
17 Rock mass quality rating (RMQR) 2015 General Aydan et al [33]
18 Modification of the rock mass rating system(rock Bolt Supporting factor (RSF)) 2017 Tunneling and
underground excavationsMohammadi and
Hossaini [34]
19 Anisotropic rock mass rating (ARMR) 2018 Tunneling and general Saroglou et al [35]
Advances in Civil Engineering 3
engineers about the visual impression of the rock structureincluding block and surface condition of the discontinuitiesrepresented by joint characteristics (roughness and alter-ation) and providing reliable data in the form of rock massstrength properties which are used as input parameters fornumerical analysis or closed-form solutions[41 46 50 55 62] +e GSI classification system has gainedwide acceptance as an empirical tool for estimating thestrength and deformation characteristics of heavily jointedrock masses [11] +e development process of the GSI wasstarted in Toronto Canada taking input about engineeringgeology from David Wood by Hoek et al [41 44] +e GSIwas further linked with Hoek and Brown failure criterion toestimate rock mass properties in true spirit in 1994 aspresented by Hoek and Kaiser in 1995 [29] and Hoek andBrown in 1997 +e GSI system classifies the rock massenvironment into five categories ranging from very good tovery poor based on the field observations about the rockmass structure ranging from blocky to disintegrated(poorly interlocked and heavily broken rock mass)[43 61 63] +e numeric value of the GSI is estimated fromthe quantified diagonal lines ranging from 10 to 80 andhaving an interval of 5 +e quality of rock mass is dividedinto five categories and four rock mass structure domains+e basic GSI chart is presented in Figure 2
4 Modifications in the GSI
+e GSI rock mass classification system was updated andmodified from time to time keeping in view its applicabilityand the mode of assessment of rock mass and estimation ofrock mass strength properties +e details of the review ofthe improvements and modification in the GSI system madeby different researchers are described in the followingparagraphs
Hoek in 1994 [63] linked the Hoek and Brown failurecriterion and its constants with each surface quality of rockmass of the basic GSI chart in order to achieve the primaryestimating objective of the GSI system for each rock massstructure domain as presented in Figure 3 without changingthe rock mass structure domain and quality in the basic GSIchart +e estimating parameters for each rock mass qualityand structure domain includes Hoek and Brown constants(s a) rock mass deformation modulus (Em) and Poissonrsquosratio (v)
Hoek et al in 1998 [43] made an extension in the GSIclassification system to accommodate the weakest AthensSchist rock masses +e foliatedlaminatedsheared rockmass category of the said area has been added to the basicGSI chart to represent thinly laminated foliated andstructurally sheared weak rocks +e rock mass included inthis category is not associated with good to very good surfacequality while it is associated with poor to very poor surfacequality due to its poor rock mass quality and shear behaviorthe rock mass falls into disintegrated and foliatedlaminatedcategories having the GSI value ranging from 5 to 30 [43]+e extension in the GSI is shown in Figure 4
From 1974 to 1998 the GSI value is estimated qualita-tively based on thorough geological visual observations of
field and expertise of professionals about collecting data +edegree of correctness of the GSI value mainly depends on theexpertise of the data collector and vice versa It means that theGSI value may be estimated differently by different peoplebased on the degree of expertise [47] +erefore it was es-sential to overcome these difficulties in the estimation of theGSI value +e era of switching from qualitative to quanti-tative estimation of the GSI value started in early 1999 in orderto estimate the GSI value appropriately as possible In thisregard the first attempt was made by Sonmez and Ulusay in1999 to develop a refined quantitative numerical basis forestimating the GSI value to suggest quantities for estimationof rock mass strength as an additional tool and to apply theGSI in the stability analysis of slopes [52] During the de-velopment of the quantification of the GSI chart Sonmez andUlusay did not consider the extension made in the basic GSIchart by Hoek et al [43] For GSI quantification Sonmez andUlusay suggested structure rating (SR) based on Jv (volu-metric joint count) which describes the rock mass structureand surface condition rating (SCR) based on roughnessweathering and infilling nature of joints SCR was calculatedusing a rating of Rr Rw and Rf [21 64] +e total rating ofSCR is obtained using the following equation [52]
SCR Rr + Rw + Rf (2)
where Rr Rw and Rf represent roughness weathering andinfilling respectively
+e SCRmaximum range of the rating is 18 and the SCRaxis is divided into 18 equal parts as shown in Figure 5
Jv is estimated from a joint survey about block size usingequations (3) and (4) [52]
Jv N1
L1+
N2
L2+
N3
L3+ middot middot middot +
Nn
Ln
(3)
Jv 1S1
+1S2
+1S3
+ middot middot middot +1Sn
(4)
where S is joint spacing N is the number of joints alongscanline L is the scanline length and n is the joint setnumber Equations (3) and (4) are used when there arelimited joints available and the rock may deal as isotropicand homogenous For highly joint conditions Sonmez andUlusay suggested the following equation to be used [52]
Jv Nx
Lx
timesNy
Ly
timesNz
Lz
middot middot middot (5)
It is complicated to assess joints in all directions during ascanline joint survey therefore for easiness the rock mass isassumed as a homogenous material For this conditionequation (5) is simplified as
Jv NL
1113874 11138753 (6)
Marinos and Hoek in 2000 and 2001 thoroughlyworked on incredible jointed rock masses surrounding thetunnels in Greece and included the intactmassive categoryin the structure domain and joint characteristics in the GSI
4 Advances in Civil Engineering
chart from lithology structure and surface conditions ofrock mass discontinuities [42 45 48] as shown in Figure 6
Cai et al in 2004 [50] introduced block volume (Vb)joint condition (Jc) factor for quantifying GSI value andinsert massive block segment in structure domain of GSI asshown in Figure 7 Inserting these parameters not onlydecreased the dependency on field experience required for
estimation of GSI value but also maintained the simplicityof GSI chart developed by Marinos and Hoek in 2000 and2001 +e influence of these two parameters ie Vb and Jcwere calibrated using published data and further verified ontwo caverns as case studies based on the back analysistechnique +e block volume is determined by the fol-lowing equation
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average values to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Blocky-very well interlocked undisturbedrock mass consisting of cubical blocks formedby three orthogonal discontinuity sets
Very blocky-interlocked partially disturbedrock mass with multifaceted angular blocksformed by four or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavilybroken rock mass with a mixture of angularand rounded rock pieces
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
may
be sl
ight
ly w
eath
ered
or i
ron-
stain
edsu
rface
s
Fair
Smoo
th an
dor
mod
erat
ely w
eath
ered
and
alte
red
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces o
r com
pact
coat
ings
with
filli
ngs o
f ang
ular
fagm
ents
Very
poo
rSl
icke
nsid
ed an
d hi
ghly
wea
ther
ed su
rface
s with
so
clay
coat
ings
or fi
lling
s
Decreasing surface quality
80
70
60
50
40
30
20
10
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Figure 2 GSI basic chart [43 61 63]
Advances in Civil Engineering 5
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
the construction of underground engineering structures[11 12] based on discontinuities and physiomechanicalproperties of rock mass for capturing the real image of the insitu conditions at the preliminary stage for effective design ofengineering structures [13 14] Numbers of empiricalmethods of design have been developed based on civil andmining engineering related case studies by various re-searchers in order to ensure the curative of the complexnature of rock mass for the design of tunnels caverns rockslopes and general purposes as shown in Table 1
+ese empirical methods have wide applications in thecivil and mining fields Each empirical approach used de-fined input parameters including field observations andlaboratory testing for solving different aspects of rock en-gineering projects [36 37] In the aforementioned empiricalmethods RMR and Q-system are widely used classificationsystems for various purposes in rock engineering andconsidered as a base for the development of other classifi-cation systems [28 33 38]
+e rock mass rating (RMR) and tunneling quality index(Q-system) used rock quality designation (RQD) as an in-tegral input parameter in the evaluation of rock mass qualityfor the design of a support system for underground struc-tures [39] +e value of RQD is determined by core drillingwhich is a very tedious procedure [17] +e procedure iscost-effective and the value of RQD for a very weak rock isusually considered zero [17] +is condition leads to anestimation of the inappropriate value of RMR and Q-systemfor the classification of rock mass [40] +e RMR and
Q-system classification systems are mostly used for simplerock mass behavior and these are less reliable to be used insqueezing defined structural failures rock bursting undervery high-stress conditions and estimation of rock massproperties for numerical modeling and less helpful ingaining information about the sequential installation ofsupports for controlling of progressive failure in weak rockmass conditions during tunneling construction [41]Keeping in view of the above weaknesses of RMR andQ-system the GSI rock mass classification system was de-veloped After development of the GSI system differentresearchers worldwide have conducted research studiesabout the various aspects of the GSI system to accommodateand evaluate weakest jointed and heterogeneous rock massenvironment for design of engineering structures and nu-merical modeling [11 13 41ndash57] +e research conducted sofar on GSI is not comprehensive reviewed in a systematicway therefore it is very important to carry out a systematicand comprehensive review of the research conducted aboutthe various aspects of GSI system in order to provide a betterunderstanding to field professionals (geologists mining andcivil engineers) about the GSI system and its application asan empirical classification and rock mass property estima-tion tool for an appropriate design of engineering structuresand numerical modeling
In this study a review of 19 years of research conductedby different researchers about the various aspects of the GSIis presented in a systematic way +e review about the GSI isdivided into four categories ie origination modifications
Siteinvest-igation
Use ofpre-existing
standardmethods
Precedent typeanalyses and
modifications
Analyticalmethods
stress-based
Rock massclassificationRMR Q GSI
Basicnumerical
methods FEMBEM DEM
hybrid
Extendednumericalmethods
fully coupledmodels
Level 11 1 mapping
Level 2Not 1 1 mapping
Integratedsystem
approachesinternet-based
Databaseexpert
systems andother systemapproaches
Design based on back analysisDesign based on forward analysis
Construction
Method A Method B Method C Method D
Objective
Figure 1 Summary of different methods for the solution of rock engineering problems [4]
2 Advances in Civil Engineering
applications and limitations in order to provide a betterunderstanding to field professionals (geologists mining andcivil engineers) about the qualitative and quantitative esti-mation of the GSI for assessment of rock mass and use theGSI as an empirical estimating tool for an appropriate designof engineering structures in rock mass environments
2 Origination of the Geological StrengthIndex (GSI)
After reviewing the empirical methods of design as pre-sented in Table 1 the following grounds are identified for thedevelopment of the GSI
(a) RQD is an integral part of RMR and Q-system Inweak rock mass conditions RQD value is taken aszero using Deree et al procedure and equation (1)[58] +is condition leads to an estimation of theinappropriate value of RMR and Q-system for rockmass classification [40]
(b) RMR Q-system and other rock mass classificationsystems are inconvenient to provide information aboutthe requirement of the support system in complextunneling conditions for controlling progressive failureand design of proper sequentially installed temporarysupports and on the contrary the numerical methodsare considered very effective to give detail informationabout this situation +e numerical methods used thephysiomechanical and deformation properties of rockmass surrounding the tunnel as input parameters
+ese properties of the rock mass are difficult tomeasure in the field due to uncertain rock massconditions and challenging procedure [7 40 58ndash60]Researchers are now focused on estimating the rockmass properties rather than finding in the field [61] Inthis regard the first attempt was made to estimate therock mass properties by Hoek and Brown in 1980 [61]Hoek and Brown developed failure criteria to estimatethe rock mass properties It was realized after thedevelopment of Hoek and Brownrsquos failure criteria thatthis failure criterion should be linked with actualgeological observations of the field to gain its practicalvalue [41]+us in this context the development of thegeological strength index (GSI) was started in Torontowith taking engineering geology as input from DavidWood in 1992 by Hoek et al to classify weak andjointed rock mass environments
RQD 1113944(length of core piecesgt 10 cm length)
times100
total length of core run
(1)
3 Geological Strength Index (GSI)
+e geological strength index (GSI) is a rock mass char-acterization tool developed for the design of tunnelscaverns and other underground structures based on fieldobservations including geological data about rock massinputs from qualified and expertise field geologists
Table 1 Empirical methods for rock mass classification and design of engineering structures in rock mass
S no Empirical method Year Applications Authors Reference1 Rock load 1946 Tunnels with steel support Terzaghi [15]2 Stand-up time 1958 Tunneling Lauffer [16]3 Rock quality designation (RQD) 1963 1966 Tunneling Deere and Miller [17 18]4 Rock structure rating (RSR) 1972 Tunneling Wickham et al [19]
5 Rock mass rating (RMR) 1973 (modifications1989) Tunnels mines slopes Bieniawski [20 21]
6 Tunneling quality index (Q) 1974 (lastmodification 2002)
Tunnels minesfoundations Barton et al [22]
7 New Austrian tunneling method (NATM) 1974 Tunneling Pacher andRabcewicz [23]
8 Size strength classification 1975 Tunneling Franklin [24]9 Basic geotechnical classification (BGC) 1981 General ISRM [25]10 Rock mass strength (RMS) 1982 Metal mining Stille and Groth [26]
11 Unified rock mass classification system(URCS) 1984 General Williamson and
Kuhn [27]
12 Rock mass index (RMi) 1996 Tunneling Palmstrom [28]
13 Geological strength index (GSI) 1997 All undergroundexcavations Hoek and Kaiser [29]
14 Rock tunneling quality index by TBMexcavation (QTBM)
1999 TBM tunnels Barton [30]
15 Continuous rock mass rating (CRMR) 2003 General Sen and Bahaaeldin [31]
16 Rock mass excitability (RME) 2006 TBM tunnels Bieniawski vonPrein et al [32]
17 Rock mass quality rating (RMQR) 2015 General Aydan et al [33]
18 Modification of the rock mass rating system(rock Bolt Supporting factor (RSF)) 2017 Tunneling and
underground excavationsMohammadi and
Hossaini [34]
19 Anisotropic rock mass rating (ARMR) 2018 Tunneling and general Saroglou et al [35]
Advances in Civil Engineering 3
engineers about the visual impression of the rock structureincluding block and surface condition of the discontinuitiesrepresented by joint characteristics (roughness and alter-ation) and providing reliable data in the form of rock massstrength properties which are used as input parameters fornumerical analysis or closed-form solutions[41 46 50 55 62] +e GSI classification system has gainedwide acceptance as an empirical tool for estimating thestrength and deformation characteristics of heavily jointedrock masses [11] +e development process of the GSI wasstarted in Toronto Canada taking input about engineeringgeology from David Wood by Hoek et al [41 44] +e GSIwas further linked with Hoek and Brown failure criterion toestimate rock mass properties in true spirit in 1994 aspresented by Hoek and Kaiser in 1995 [29] and Hoek andBrown in 1997 +e GSI system classifies the rock massenvironment into five categories ranging from very good tovery poor based on the field observations about the rockmass structure ranging from blocky to disintegrated(poorly interlocked and heavily broken rock mass)[43 61 63] +e numeric value of the GSI is estimated fromthe quantified diagonal lines ranging from 10 to 80 andhaving an interval of 5 +e quality of rock mass is dividedinto five categories and four rock mass structure domains+e basic GSI chart is presented in Figure 2
4 Modifications in the GSI
+e GSI rock mass classification system was updated andmodified from time to time keeping in view its applicabilityand the mode of assessment of rock mass and estimation ofrock mass strength properties +e details of the review ofthe improvements and modification in the GSI system madeby different researchers are described in the followingparagraphs
Hoek in 1994 [63] linked the Hoek and Brown failurecriterion and its constants with each surface quality of rockmass of the basic GSI chart in order to achieve the primaryestimating objective of the GSI system for each rock massstructure domain as presented in Figure 3 without changingthe rock mass structure domain and quality in the basic GSIchart +e estimating parameters for each rock mass qualityand structure domain includes Hoek and Brown constants(s a) rock mass deformation modulus (Em) and Poissonrsquosratio (v)
Hoek et al in 1998 [43] made an extension in the GSIclassification system to accommodate the weakest AthensSchist rock masses +e foliatedlaminatedsheared rockmass category of the said area has been added to the basicGSI chart to represent thinly laminated foliated andstructurally sheared weak rocks +e rock mass included inthis category is not associated with good to very good surfacequality while it is associated with poor to very poor surfacequality due to its poor rock mass quality and shear behaviorthe rock mass falls into disintegrated and foliatedlaminatedcategories having the GSI value ranging from 5 to 30 [43]+e extension in the GSI is shown in Figure 4
From 1974 to 1998 the GSI value is estimated qualita-tively based on thorough geological visual observations of
field and expertise of professionals about collecting data +edegree of correctness of the GSI value mainly depends on theexpertise of the data collector and vice versa It means that theGSI value may be estimated differently by different peoplebased on the degree of expertise [47] +erefore it was es-sential to overcome these difficulties in the estimation of theGSI value +e era of switching from qualitative to quanti-tative estimation of the GSI value started in early 1999 in orderto estimate the GSI value appropriately as possible In thisregard the first attempt was made by Sonmez and Ulusay in1999 to develop a refined quantitative numerical basis forestimating the GSI value to suggest quantities for estimationof rock mass strength as an additional tool and to apply theGSI in the stability analysis of slopes [52] During the de-velopment of the quantification of the GSI chart Sonmez andUlusay did not consider the extension made in the basic GSIchart by Hoek et al [43] For GSI quantification Sonmez andUlusay suggested structure rating (SR) based on Jv (volu-metric joint count) which describes the rock mass structureand surface condition rating (SCR) based on roughnessweathering and infilling nature of joints SCR was calculatedusing a rating of Rr Rw and Rf [21 64] +e total rating ofSCR is obtained using the following equation [52]
SCR Rr + Rw + Rf (2)
where Rr Rw and Rf represent roughness weathering andinfilling respectively
+e SCRmaximum range of the rating is 18 and the SCRaxis is divided into 18 equal parts as shown in Figure 5
Jv is estimated from a joint survey about block size usingequations (3) and (4) [52]
Jv N1
L1+
N2
L2+
N3
L3+ middot middot middot +
Nn
Ln
(3)
Jv 1S1
+1S2
+1S3
+ middot middot middot +1Sn
(4)
where S is joint spacing N is the number of joints alongscanline L is the scanline length and n is the joint setnumber Equations (3) and (4) are used when there arelimited joints available and the rock may deal as isotropicand homogenous For highly joint conditions Sonmez andUlusay suggested the following equation to be used [52]
Jv Nx
Lx
timesNy
Ly
timesNz
Lz
middot middot middot (5)
It is complicated to assess joints in all directions during ascanline joint survey therefore for easiness the rock mass isassumed as a homogenous material For this conditionequation (5) is simplified as
Jv NL
1113874 11138753 (6)
Marinos and Hoek in 2000 and 2001 thoroughlyworked on incredible jointed rock masses surrounding thetunnels in Greece and included the intactmassive categoryin the structure domain and joint characteristics in the GSI
4 Advances in Civil Engineering
chart from lithology structure and surface conditions ofrock mass discontinuities [42 45 48] as shown in Figure 6
Cai et al in 2004 [50] introduced block volume (Vb)joint condition (Jc) factor for quantifying GSI value andinsert massive block segment in structure domain of GSI asshown in Figure 7 Inserting these parameters not onlydecreased the dependency on field experience required for
estimation of GSI value but also maintained the simplicityof GSI chart developed by Marinos and Hoek in 2000 and2001 +e influence of these two parameters ie Vb and Jcwere calibrated using published data and further verified ontwo caverns as case studies based on the back analysistechnique +e block volume is determined by the fol-lowing equation
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average values to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Blocky-very well interlocked undisturbedrock mass consisting of cubical blocks formedby three orthogonal discontinuity sets
Very blocky-interlocked partially disturbedrock mass with multifaceted angular blocksformed by four or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavilybroken rock mass with a mixture of angularand rounded rock pieces
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
may
be sl
ight
ly w
eath
ered
or i
ron-
stain
edsu
rface
s
Fair
Smoo
th an
dor
mod
erat
ely w
eath
ered
and
alte
red
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces o
r com
pact
coat
ings
with
filli
ngs o
f ang
ular
fagm
ents
Very
poo
rSl
icke
nsid
ed an
d hi
ghly
wea
ther
ed su
rface
s with
so
clay
coat
ings
or fi
lling
s
Decreasing surface quality
80
70
60
50
40
30
20
10
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Figure 2 GSI basic chart [43 61 63]
Advances in Civil Engineering 5
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
applications and limitations in order to provide a betterunderstanding to field professionals (geologists mining andcivil engineers) about the qualitative and quantitative esti-mation of the GSI for assessment of rock mass and use theGSI as an empirical estimating tool for an appropriate designof engineering structures in rock mass environments
2 Origination of the Geological StrengthIndex (GSI)
After reviewing the empirical methods of design as pre-sented in Table 1 the following grounds are identified for thedevelopment of the GSI
(a) RQD is an integral part of RMR and Q-system Inweak rock mass conditions RQD value is taken aszero using Deree et al procedure and equation (1)[58] +is condition leads to an estimation of theinappropriate value of RMR and Q-system for rockmass classification [40]
(b) RMR Q-system and other rock mass classificationsystems are inconvenient to provide information aboutthe requirement of the support system in complextunneling conditions for controlling progressive failureand design of proper sequentially installed temporarysupports and on the contrary the numerical methodsare considered very effective to give detail informationabout this situation +e numerical methods used thephysiomechanical and deformation properties of rockmass surrounding the tunnel as input parameters
+ese properties of the rock mass are difficult tomeasure in the field due to uncertain rock massconditions and challenging procedure [7 40 58ndash60]Researchers are now focused on estimating the rockmass properties rather than finding in the field [61] Inthis regard the first attempt was made to estimate therock mass properties by Hoek and Brown in 1980 [61]Hoek and Brown developed failure criteria to estimatethe rock mass properties It was realized after thedevelopment of Hoek and Brownrsquos failure criteria thatthis failure criterion should be linked with actualgeological observations of the field to gain its practicalvalue [41]+us in this context the development of thegeological strength index (GSI) was started in Torontowith taking engineering geology as input from DavidWood in 1992 by Hoek et al to classify weak andjointed rock mass environments
RQD 1113944(length of core piecesgt 10 cm length)
times100
total length of core run
(1)
3 Geological Strength Index (GSI)
+e geological strength index (GSI) is a rock mass char-acterization tool developed for the design of tunnelscaverns and other underground structures based on fieldobservations including geological data about rock massinputs from qualified and expertise field geologists
Table 1 Empirical methods for rock mass classification and design of engineering structures in rock mass
S no Empirical method Year Applications Authors Reference1 Rock load 1946 Tunnels with steel support Terzaghi [15]2 Stand-up time 1958 Tunneling Lauffer [16]3 Rock quality designation (RQD) 1963 1966 Tunneling Deere and Miller [17 18]4 Rock structure rating (RSR) 1972 Tunneling Wickham et al [19]
5 Rock mass rating (RMR) 1973 (modifications1989) Tunnels mines slopes Bieniawski [20 21]
6 Tunneling quality index (Q) 1974 (lastmodification 2002)
Tunnels minesfoundations Barton et al [22]
7 New Austrian tunneling method (NATM) 1974 Tunneling Pacher andRabcewicz [23]
8 Size strength classification 1975 Tunneling Franklin [24]9 Basic geotechnical classification (BGC) 1981 General ISRM [25]10 Rock mass strength (RMS) 1982 Metal mining Stille and Groth [26]
11 Unified rock mass classification system(URCS) 1984 General Williamson and
Kuhn [27]
12 Rock mass index (RMi) 1996 Tunneling Palmstrom [28]
13 Geological strength index (GSI) 1997 All undergroundexcavations Hoek and Kaiser [29]
14 Rock tunneling quality index by TBMexcavation (QTBM)
1999 TBM tunnels Barton [30]
15 Continuous rock mass rating (CRMR) 2003 General Sen and Bahaaeldin [31]
16 Rock mass excitability (RME) 2006 TBM tunnels Bieniawski vonPrein et al [32]
17 Rock mass quality rating (RMQR) 2015 General Aydan et al [33]
18 Modification of the rock mass rating system(rock Bolt Supporting factor (RSF)) 2017 Tunneling and
underground excavationsMohammadi and
Hossaini [34]
19 Anisotropic rock mass rating (ARMR) 2018 Tunneling and general Saroglou et al [35]
Advances in Civil Engineering 3
engineers about the visual impression of the rock structureincluding block and surface condition of the discontinuitiesrepresented by joint characteristics (roughness and alter-ation) and providing reliable data in the form of rock massstrength properties which are used as input parameters fornumerical analysis or closed-form solutions[41 46 50 55 62] +e GSI classification system has gainedwide acceptance as an empirical tool for estimating thestrength and deformation characteristics of heavily jointedrock masses [11] +e development process of the GSI wasstarted in Toronto Canada taking input about engineeringgeology from David Wood by Hoek et al [41 44] +e GSIwas further linked with Hoek and Brown failure criterion toestimate rock mass properties in true spirit in 1994 aspresented by Hoek and Kaiser in 1995 [29] and Hoek andBrown in 1997 +e GSI system classifies the rock massenvironment into five categories ranging from very good tovery poor based on the field observations about the rockmass structure ranging from blocky to disintegrated(poorly interlocked and heavily broken rock mass)[43 61 63] +e numeric value of the GSI is estimated fromthe quantified diagonal lines ranging from 10 to 80 andhaving an interval of 5 +e quality of rock mass is dividedinto five categories and four rock mass structure domains+e basic GSI chart is presented in Figure 2
4 Modifications in the GSI
+e GSI rock mass classification system was updated andmodified from time to time keeping in view its applicabilityand the mode of assessment of rock mass and estimation ofrock mass strength properties +e details of the review ofthe improvements and modification in the GSI system madeby different researchers are described in the followingparagraphs
Hoek in 1994 [63] linked the Hoek and Brown failurecriterion and its constants with each surface quality of rockmass of the basic GSI chart in order to achieve the primaryestimating objective of the GSI system for each rock massstructure domain as presented in Figure 3 without changingthe rock mass structure domain and quality in the basic GSIchart +e estimating parameters for each rock mass qualityand structure domain includes Hoek and Brown constants(s a) rock mass deformation modulus (Em) and Poissonrsquosratio (v)
Hoek et al in 1998 [43] made an extension in the GSIclassification system to accommodate the weakest AthensSchist rock masses +e foliatedlaminatedsheared rockmass category of the said area has been added to the basicGSI chart to represent thinly laminated foliated andstructurally sheared weak rocks +e rock mass included inthis category is not associated with good to very good surfacequality while it is associated with poor to very poor surfacequality due to its poor rock mass quality and shear behaviorthe rock mass falls into disintegrated and foliatedlaminatedcategories having the GSI value ranging from 5 to 30 [43]+e extension in the GSI is shown in Figure 4
From 1974 to 1998 the GSI value is estimated qualita-tively based on thorough geological visual observations of
field and expertise of professionals about collecting data +edegree of correctness of the GSI value mainly depends on theexpertise of the data collector and vice versa It means that theGSI value may be estimated differently by different peoplebased on the degree of expertise [47] +erefore it was es-sential to overcome these difficulties in the estimation of theGSI value +e era of switching from qualitative to quanti-tative estimation of the GSI value started in early 1999 in orderto estimate the GSI value appropriately as possible In thisregard the first attempt was made by Sonmez and Ulusay in1999 to develop a refined quantitative numerical basis forestimating the GSI value to suggest quantities for estimationof rock mass strength as an additional tool and to apply theGSI in the stability analysis of slopes [52] During the de-velopment of the quantification of the GSI chart Sonmez andUlusay did not consider the extension made in the basic GSIchart by Hoek et al [43] For GSI quantification Sonmez andUlusay suggested structure rating (SR) based on Jv (volu-metric joint count) which describes the rock mass structureand surface condition rating (SCR) based on roughnessweathering and infilling nature of joints SCR was calculatedusing a rating of Rr Rw and Rf [21 64] +e total rating ofSCR is obtained using the following equation [52]
SCR Rr + Rw + Rf (2)
where Rr Rw and Rf represent roughness weathering andinfilling respectively
+e SCRmaximum range of the rating is 18 and the SCRaxis is divided into 18 equal parts as shown in Figure 5
Jv is estimated from a joint survey about block size usingequations (3) and (4) [52]
Jv N1
L1+
N2
L2+
N3
L3+ middot middot middot +
Nn
Ln
(3)
Jv 1S1
+1S2
+1S3
+ middot middot middot +1Sn
(4)
where S is joint spacing N is the number of joints alongscanline L is the scanline length and n is the joint setnumber Equations (3) and (4) are used when there arelimited joints available and the rock may deal as isotropicand homogenous For highly joint conditions Sonmez andUlusay suggested the following equation to be used [52]
Jv Nx
Lx
timesNy
Ly
timesNz
Lz
middot middot middot (5)
It is complicated to assess joints in all directions during ascanline joint survey therefore for easiness the rock mass isassumed as a homogenous material For this conditionequation (5) is simplified as
Jv NL
1113874 11138753 (6)
Marinos and Hoek in 2000 and 2001 thoroughlyworked on incredible jointed rock masses surrounding thetunnels in Greece and included the intactmassive categoryin the structure domain and joint characteristics in the GSI
4 Advances in Civil Engineering
chart from lithology structure and surface conditions ofrock mass discontinuities [42 45 48] as shown in Figure 6
Cai et al in 2004 [50] introduced block volume (Vb)joint condition (Jc) factor for quantifying GSI value andinsert massive block segment in structure domain of GSI asshown in Figure 7 Inserting these parameters not onlydecreased the dependency on field experience required for
estimation of GSI value but also maintained the simplicityof GSI chart developed by Marinos and Hoek in 2000 and2001 +e influence of these two parameters ie Vb and Jcwere calibrated using published data and further verified ontwo caverns as case studies based on the back analysistechnique +e block volume is determined by the fol-lowing equation
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average values to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Blocky-very well interlocked undisturbedrock mass consisting of cubical blocks formedby three orthogonal discontinuity sets
Very blocky-interlocked partially disturbedrock mass with multifaceted angular blocksformed by four or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavilybroken rock mass with a mixture of angularand rounded rock pieces
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
may
be sl
ight
ly w
eath
ered
or i
ron-
stain
edsu
rface
s
Fair
Smoo
th an
dor
mod
erat
ely w
eath
ered
and
alte
red
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces o
r com
pact
coat
ings
with
filli
ngs o
f ang
ular
fagm
ents
Very
poo
rSl
icke
nsid
ed an
d hi
ghly
wea
ther
ed su
rface
s with
so
clay
coat
ings
or fi
lling
s
Decreasing surface quality
80
70
60
50
40
30
20
10
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Figure 2 GSI basic chart [43 61 63]
Advances in Civil Engineering 5
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
engineers about the visual impression of the rock structureincluding block and surface condition of the discontinuitiesrepresented by joint characteristics (roughness and alter-ation) and providing reliable data in the form of rock massstrength properties which are used as input parameters fornumerical analysis or closed-form solutions[41 46 50 55 62] +e GSI classification system has gainedwide acceptance as an empirical tool for estimating thestrength and deformation characteristics of heavily jointedrock masses [11] +e development process of the GSI wasstarted in Toronto Canada taking input about engineeringgeology from David Wood by Hoek et al [41 44] +e GSIwas further linked with Hoek and Brown failure criterion toestimate rock mass properties in true spirit in 1994 aspresented by Hoek and Kaiser in 1995 [29] and Hoek andBrown in 1997 +e GSI system classifies the rock massenvironment into five categories ranging from very good tovery poor based on the field observations about the rockmass structure ranging from blocky to disintegrated(poorly interlocked and heavily broken rock mass)[43 61 63] +e numeric value of the GSI is estimated fromthe quantified diagonal lines ranging from 10 to 80 andhaving an interval of 5 +e quality of rock mass is dividedinto five categories and four rock mass structure domains+e basic GSI chart is presented in Figure 2
4 Modifications in the GSI
+e GSI rock mass classification system was updated andmodified from time to time keeping in view its applicabilityand the mode of assessment of rock mass and estimation ofrock mass strength properties +e details of the review ofthe improvements and modification in the GSI system madeby different researchers are described in the followingparagraphs
Hoek in 1994 [63] linked the Hoek and Brown failurecriterion and its constants with each surface quality of rockmass of the basic GSI chart in order to achieve the primaryestimating objective of the GSI system for each rock massstructure domain as presented in Figure 3 without changingthe rock mass structure domain and quality in the basic GSIchart +e estimating parameters for each rock mass qualityand structure domain includes Hoek and Brown constants(s a) rock mass deformation modulus (Em) and Poissonrsquosratio (v)
Hoek et al in 1998 [43] made an extension in the GSIclassification system to accommodate the weakest AthensSchist rock masses +e foliatedlaminatedsheared rockmass category of the said area has been added to the basicGSI chart to represent thinly laminated foliated andstructurally sheared weak rocks +e rock mass included inthis category is not associated with good to very good surfacequality while it is associated with poor to very poor surfacequality due to its poor rock mass quality and shear behaviorthe rock mass falls into disintegrated and foliatedlaminatedcategories having the GSI value ranging from 5 to 30 [43]+e extension in the GSI is shown in Figure 4
From 1974 to 1998 the GSI value is estimated qualita-tively based on thorough geological visual observations of
field and expertise of professionals about collecting data +edegree of correctness of the GSI value mainly depends on theexpertise of the data collector and vice versa It means that theGSI value may be estimated differently by different peoplebased on the degree of expertise [47] +erefore it was es-sential to overcome these difficulties in the estimation of theGSI value +e era of switching from qualitative to quanti-tative estimation of the GSI value started in early 1999 in orderto estimate the GSI value appropriately as possible In thisregard the first attempt was made by Sonmez and Ulusay in1999 to develop a refined quantitative numerical basis forestimating the GSI value to suggest quantities for estimationof rock mass strength as an additional tool and to apply theGSI in the stability analysis of slopes [52] During the de-velopment of the quantification of the GSI chart Sonmez andUlusay did not consider the extension made in the basic GSIchart by Hoek et al [43] For GSI quantification Sonmez andUlusay suggested structure rating (SR) based on Jv (volu-metric joint count) which describes the rock mass structureand surface condition rating (SCR) based on roughnessweathering and infilling nature of joints SCR was calculatedusing a rating of Rr Rw and Rf [21 64] +e total rating ofSCR is obtained using the following equation [52]
SCR Rr + Rw + Rf (2)
where Rr Rw and Rf represent roughness weathering andinfilling respectively
+e SCRmaximum range of the rating is 18 and the SCRaxis is divided into 18 equal parts as shown in Figure 5
Jv is estimated from a joint survey about block size usingequations (3) and (4) [52]
Jv N1
L1+
N2
L2+
N3
L3+ middot middot middot +
Nn
Ln
(3)
Jv 1S1
+1S2
+1S3
+ middot middot middot +1Sn
(4)
where S is joint spacing N is the number of joints alongscanline L is the scanline length and n is the joint setnumber Equations (3) and (4) are used when there arelimited joints available and the rock may deal as isotropicand homogenous For highly joint conditions Sonmez andUlusay suggested the following equation to be used [52]
Jv Nx
Lx
timesNy
Ly
timesNz
Lz
middot middot middot (5)
It is complicated to assess joints in all directions during ascanline joint survey therefore for easiness the rock mass isassumed as a homogenous material For this conditionequation (5) is simplified as
Jv NL
1113874 11138753 (6)
Marinos and Hoek in 2000 and 2001 thoroughlyworked on incredible jointed rock masses surrounding thetunnels in Greece and included the intactmassive categoryin the structure domain and joint characteristics in the GSI
4 Advances in Civil Engineering
chart from lithology structure and surface conditions ofrock mass discontinuities [42 45 48] as shown in Figure 6
Cai et al in 2004 [50] introduced block volume (Vb)joint condition (Jc) factor for quantifying GSI value andinsert massive block segment in structure domain of GSI asshown in Figure 7 Inserting these parameters not onlydecreased the dependency on field experience required for
estimation of GSI value but also maintained the simplicityof GSI chart developed by Marinos and Hoek in 2000 and2001 +e influence of these two parameters ie Vb and Jcwere calibrated using published data and further verified ontwo caverns as case studies based on the back analysistechnique +e block volume is determined by the fol-lowing equation
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average values to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Blocky-very well interlocked undisturbedrock mass consisting of cubical blocks formedby three orthogonal discontinuity sets
Very blocky-interlocked partially disturbedrock mass with multifaceted angular blocksformed by four or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavilybroken rock mass with a mixture of angularand rounded rock pieces
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
may
be sl
ight
ly w
eath
ered
or i
ron-
stain
edsu
rface
s
Fair
Smoo
th an
dor
mod
erat
ely w
eath
ered
and
alte
red
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces o
r com
pact
coat
ings
with
filli
ngs o
f ang
ular
fagm
ents
Very
poo
rSl
icke
nsid
ed an
d hi
ghly
wea
ther
ed su
rface
s with
so
clay
coat
ings
or fi
lling
s
Decreasing surface quality
80
70
60
50
40
30
20
10
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Figure 2 GSI basic chart [43 61 63]
Advances in Civil Engineering 5
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
chart from lithology structure and surface conditions ofrock mass discontinuities [42 45 48] as shown in Figure 6
Cai et al in 2004 [50] introduced block volume (Vb)joint condition (Jc) factor for quantifying GSI value andinsert massive block segment in structure domain of GSI asshown in Figure 7 Inserting these parameters not onlydecreased the dependency on field experience required for
estimation of GSI value but also maintained the simplicityof GSI chart developed by Marinos and Hoek in 2000 and2001 +e influence of these two parameters ie Vb and Jcwere calibrated using published data and further verified ontwo caverns as case studies based on the back analysistechnique +e block volume is determined by the fol-lowing equation
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average values to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Blocky-very well interlocked undisturbedrock mass consisting of cubical blocks formedby three orthogonal discontinuity sets
Very blocky-interlocked partially disturbedrock mass with multifaceted angular blocksformed by four or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavilybroken rock mass with a mixture of angularand rounded rock pieces
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
may
be sl
ight
ly w
eath
ered
or i
ron-
stain
edsu
rface
s
Fair
Smoo
th an
dor
mod
erat
ely w
eath
ered
and
alte
red
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces o
r com
pact
coat
ings
with
filli
ngs o
f ang
ular
fagm
ents
Very
poo
rSl
icke
nsid
ed an
d hi
ghly
wea
ther
ed su
rface
s with
so
clay
coat
ings
or fi
lling
s
Decreasing surface quality
80
70
60
50
40
30
20
10
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Figure 2 GSI basic chart [43 61 63]
Advances in Civil Engineering 5
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
Vb s1 times s2 times s3
sin c1 times sin c2 times sin c3 (7)
where s is joint spacing and c is the angle between joint setsas shown in Figure 8
Due to variations in joint spacing the effect of c betweenjoint sets is relatively small +erefore the block volume canbe estimated for practical purpose as
Vb s1 times s2 times s3 (8)
Since joints are discontinued it is essential to determineequivalent block volume using the following equation
Vb s1 times s2 times s3P1 times P2 times P3
31113968 times
1sin c1 times sin c2 times sin c3
(9)
Very
goo
dVe
ry ro
ugh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
r alte
red
surfa
ces
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
or fi
lling
s con
tain
ing
angu
lar
rock
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ithso
cla
y co
atin
gs o
r filli
ngs
Surfa
ce co
nditi
on
Generalised HoekndashBrown Criteria
Structure
Blocky - very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlockedpartially disturbed rock mass withmultifaceted angular blocks formedby four or more discontinuity sets
Blockyseamy - folded andfaulted with many intersectingdiscontinuities forming angularblocks
Crushed - poorly interlockedheavily broken rock mass with amixture of angular and roundedblocks
0600190
0575000
0285
0400062
0540000
0275
0260015
0520000
02562
00800004
05300002534
0160003
05900002548
0240012
0518000
02560
0170004
0510000
02550
0120001
05600002540
0060
0552000
0320
0080
053000
0330
0170004
0510000
02550
0120001
05600002540
0080
053000
0330
0040
0601000
0310
0060
0552000
0320
0400062
0540000
0275
0290021
0524000
02565
0160003
05900002548
0070
0532500
0325
0110001
05500002538
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
mbmisα
Emν
GSI
σ1 prime= major principal effective stress at failure
σ3 prime= minor principal effective stress at failure
σc = uniaxial compressive strength of intactpieces of rock
mb s and α are constants which depend onthe composition structure and surfaceconditions of the rock mass
σ1 prime= σ3 prime+ σc (mb (σ3primeσc) + S)a
Figure 3 Hoek and Brown failure criterion linkage with the GSI [63]
6 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
Geological strength index
From the description of structure and surface conditions ofthe rock mass pick an appropriate box in this chartEstimate the average value to the geological strengthindex (GSI) from the contours Do not attempt to be tooprecise Quoting a range of GSI from 36 to 42 is morerealistic than stating that GSI = 38 It is also important torecognize that the HoekndashBrown criterion should only beapplied to rock masses where the size of individual blocksis small compared with the size of the excavation underconsideration
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
and
fresh
unw
eath
ered
surfa
ces
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stain
ed su
rface
s
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
dsu
rface
s
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ctco
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blockyndashvery well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky-interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed-folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated-poorly interlocked heavily brokenrock mass with a mixture of angular and roundedrock pieces
Foliatedlaminatedsheard-thinly laminated or foliated tectonicallysheared weak rocks closely spaced schistosityprevails over any other discontinuity set resultingin complete lack of blockiness
NA NA
10
5
20
30
40
50
60
70
80
Decreasing surface quality
Figure 4 Extension in the basic GSI chart [43]
Advances in Civil Engineering 7
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
Blocky - very well interlocked undisturbed rockmass consisting of cubical blocks formed by threeorthogonal discontinuity sets
Very blocky - interlocked partially disturbed rockmass with multifaceted angular blocks formed byfour or more discontinuity sets
Blockydisturbed - folded andor faulted withangular blocks formed by many intersectingdiscontinuity sets
Disintegrated- poorly interlocked heavily brokenrock mass with a mixture or angular and roundedrock pieces
Ver
y go
odV
ery
roug
h fr
esh
unw
eath
ered
surfa
ces
Goo
dSm
ooth
slig
htly
wea
ther
ed i
ron-
stain
edsu
rface
s
Fair
Smoo
th m
oder
atel
y w
eath
ered
or
alte
rted
surfa
ces
Poor
Slic
kens
ided
or h
ighl
y w
eath
ered
surfa
ces w
ith co
mpa
ct co
atin
gsor
filli
ngs o
f ang
ular
frag
men
ts
Ver
y po
orSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cl
ay co
atin
gs o
r filli
ngs
100
90
80
70
60
50
40
30
20
10
001 1 10 102 103 104
Volumetric joint count Jv (jointm3)
Stru
ctur
e rat
ing
(SR)
Roughnessrating (Rr)
Weatheringrating (Rw)
Infillingrating (Rf)
Veryrough
6Rough
5
None6
None6
Slightlyweathered
5Hard
lt 5mm4
Hardgt 5mm
2
Sogt 5mm
2
Sogt 5mm
0
Slightlyrough
3Moderatelyweathered
3
Highlyweathered
3
Smooth1
Slickensided0
Decomposed0
Blocky VB BD Disintegrated
100
95
90
85
80
80
75
70
70
65
60
60
55
50
50
45
40
40
35
30 30
25
2020
15
10
105
0
Stru
ctur
e rat
ing
(SR)
0123456789101112131415161718Surface condition rating (SCR)
SCR = Rr + Rw + Rf
Figure 5 Quantification of the GSI chart [52]
8 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
From the lithology structure and surfaceconditions of the discontinuities estimatethe average value of GSI Do not try tobe too precise Quoting a range from 33to 37 is more realistic than stating thatGSI = 35 Note that the table does notapply to structurally controlled failuresWhere weak planar structural planes arepresent in an unfavourable orientationwith respect to the excavation face thesewill dominate the rock mass behavioure shear strength of surfaces in rocksthat are prone to deterioration as a resultof changes in moisture content will bereduced is water is present Whenworking with rocks in the fair to very poorcategories a shi to the right may bemade for wet conditions Water pressureis dealt by effective stress analysis
Geological strength index forjointed rocks [49]
Structure
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ings
with
filli
ngs o
r ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
Decreasing surface quality
Dec
reas
ing
inte
rlock
ing
of ro
ck p
iece
s
Blocky - well interlocked undisturbedrock mass consisting of cubical blocks formedby three intersecting discontinuity sets
Very blocky - interlockedpartially disturbed mass withmultifaceted angular blocksformed by 4 or more joint sets
Blockydisturbedseamy-folded with angular blocks formedby many intersecting discontinuity sets Persistence of bedding planes or schistosity
Disintegrated - poorly interlockedheavily broken rock masswith mixture of angular androunded rock pieces
Laminatedsheared - Lackof blockiness due to close spacingof weak schistosity or shear planes NA NA
NA NA90
80
70
60
50
40
30
20
10
Intact or massive - intactrock specimens or massive insitu rock with few widely spaceddiscontinuities
Figure 6 GSI chart for jointed rock mass [42 45 48]
Advances in Civil Engineering 9
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed o
ral
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
coat
ing
or fi
lling
s of a
ngul
ar fr
agm
ents
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
y co
atin
gs o
r filli
ngs
10E + 6
1E + 6(1m3)
100E + 3
10E + 3
1000(1dm3)
100
10
1
5
10
0112 45 17 067 025 01
Joint condition factor Jc
Joint or block wall condition
GSI
Block size
Massive - very well interlockedundisturbed rock mass blocks formedby three or less discontinuity setswith very wide joint spacingJoint spacing gt 100cm
Blocky-very well interlockedundisturbed rock mass consistingof cubical blocks formed by threeorthogonal discontinuity setsJoint spacing 30ndash100cm
Very blocky-interlocked partiallydisturbed rock mass with multifacetedangular blocks formed by four or morediscontinuity setsJoint spacing 10ndash30cm
Blockydisturbed-folded andorfaulted with angular blocks formedby many intersecting discontinuity setsJoint spacing 3ndash10cm
Disintegrated poorly interlockedheavily broken rock mass with amixture or angularand rounded rock piecesJoint spacing lt 3cm
Foliatedlaminatedsheared-thinlylaminated or foliated tectonically shearedweak rock closely spaced schistosityprevails over any other discontinuity setresulting in complete lack of blockinessJoint spacing lt 1cm
1cm
10cm
100cm
30cm
95
85
75
65
5570
60
50
40
30
20
25
35
451
90
80
4
3
15
Predicted GSI
Back calculated GSI
5
2
Britt
le fai
lure z
one
Britt
le fai
lure z
one150
9080706050
40
20
5
3
2
NA NA
Bloc
k vo
lum
e Vb (
cm3 )
Figure 7 GSI quantification chart [49]
10 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
where P is the joint persistent factor+e joint condition factor similar to the factor used by
Palmstrom [28] is determined by the following equation
Jc Js times Jw
Ja
(10)
where Js Jw and Ja are joint spacing roughness and al-teration ratings
Hoek et al in 2005 [64] incorporated weak hetero-geneous rock masses and lithological variability of a rockmass Hoek et al developed two charts for confined andfissile molaissic rock masses for tunnels and surfaceexcavations
Hoek et al in 2013 inserted scales at the x-axis and y-axis represented by A and B respectively in the GSI chartdeveloped by Hoek and Marinos 2000 Scale A is repre-senting the surface quality of rock mass having a range from0 to 45 divided into 5 divisions at an interval of 9 Scale B isrepresenting block interlocking and structure domains of therock mass having a range from 0 to 50 divided into 5 di-visions at an interval of 10 [13]
Furthermore Marinos and Hoek defined scale A by 15JCond89 and scale B is defined as RQD2 in the basic GSIchart as shown in Figure 9
5 Applications of the GSI as an EmpiricalEstimation Tool for GeomechanicalProperties of Rock Mass
+eGSI system is explicitly used in estimations of rock massstrength and deformation properties based on Hoek andBrown failure criterion for numerical modeling and analysisof projects in rock engineering [8 9 11 22 39ndash41 43 4749ndash51 53 54 61 64ndash73] +e exceptionality of the GSI overother empirical methods (RMR Q-system etc) is that thisempirical method determined the rock mass strength anddeformation properties of weak to very weak rock massenvironment and addressed the rock heterogeneity in thebest way +e details of the GSI as an empirical estimationtool are discussed in the following
51 Estimation of the Generalized H-B Failure CriterionConstants andMohrndashCoulomb Failure Constants Hoek andBrown in 1980 [61 74 75] proposed the nonlinear failurecriterion as given in equation (11) for estimation of intactrock strength based on a wide range of triaxial tests on intactrock samples
σ1 σ3 +
mCoσ3 + sCo21113969
(11)
where σ1 and σ3 are major and minor principal stressesrespectively Co is the uniaxial compressive strength and mand s are Hoek and Brown constants Equation (11) can alsodenoted as
σ1 σ3 + σci mσ3σci
+ s1113888 1113889
05
(12)
+e Hoek and Brown failure criterion is updated fromtime to time in accordance to experience gained and toaddress certain practical limitations [29 44 61 76] Hoekand Brown developed a generalized failure criterion alsoknown as 2002 edition to estimate jointed rock massstrength as given in the following equation [11 13 4248 49 76 77]
σ1 σ3 + σc mb
σ3σc
+ s1113888 1113889
a
(13)
where mb s and a are material constants Hoek and Brownconstants are determined by equations (14)ndash(16) respec-tively when the GSI value and D are known [78 79] +eother empirical methods are used to determine the Hoek andBrown constants
mb mi expGSI minus 10028 minus 14D
1113874 1113875a
(14)
s expGSI minus 1009 minus 3D
1113874 1113875 (15)
a 05 +16
eminus GSI15
minus eminus 203
1113872 1113873 (16)
whereD is a disturbance factor which depends on the natureof the blast and mi is the intact rock constant
+e HoekndashBrown constants mb s and a for jointed rockmass are determined using the following equations withoutdisturbance factor (D) or when the D value is zero (controlblast of excavation by TBM) [52]
When the GSI value is estimated then the followingequation is used to determine the value of mb
mb mi expGSI minus 100
281113874 1113875 (17)
When GSIgt 25 then s is calculated from the followingequation as
s expGSI minus 100
91113874 1113875 (18)
For good quality of rock mass
Joint set 1
Joint set 2
Joint set 3
S3
S2
S1
γ2
γ1
γ3
Figure 8 Joint sets and angle between joint sets [49]
Advances in Civil Engineering 11
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
a 05 (19)
When GSIlt 25 then the rock mass is of poor qualityand
s 0 (20)
And the value of a is determined by the followingequation
a 065 minusGSI200
(21)
+eMohrndashCoulomb failure criterion is also called as thelinear failure criterion which is used in estimation ofstrength of rocks for design purpose +e constants iecohesion (Cm) and angle of internal friction (φm) or shearstrength parameters are empirically estimated using
Geological strength index (GSI)for jointed blocky rock masses
From the lithology structure and observeddiscontinuity surface conditions estimate theaverage GSI based on the descriptions inthe row and column headings Alternativelyfrom logged RQD values and joint conditionratings (from Bieniawski 1989) estimateGSI = 15 JCond89 + RQD2 based on thescales attached to the chart axes
For intact or massive rock with GSI gt 75check for brittle spalling potential Forsparsely jointed rock with GSI gt 75 failurewill be controlled by structurally defined blocksor wedges e HoekndashBrown criterion shouldnot be used for either of these conditions
is chart applies to tunnels of about 10mspan and slopes lt 20m high For largerCaverns and slopes consider reducing GSIto account for decreasing block interlocking
Blocky-well interlockedundisturbed rock mass madeup of cubical blocks formed bythree sets of intersecting joints
Very blocky-interlockedpartially disturbed rock massmultifaceted angular blocksformed by 4 or more joint sets
Blocky disturbedseamy-folded with angular blocks formedby many intersecting joint setsPersistence of bedding planes orschistosity
Disintegrated-poorly inter-locked heavily broken rock masswith mixture of angular and roundedrock pieces
Structure Decreasing surface quality40
4015 JCond89
35
35
30
30
25
25
20
20
15
15
10
10
5
50
045
Dec
reas
ing
inte
rlock
ing
Surfa
ce co
nditi
ons
Very
goo
dVe
ry ro
ugh
fres
h un
wea
ther
ed su
rface
s
Goo
dRo
ugh
slig
htly
wea
ther
ed i
ron-
stai
ned
surfa
ces
Fair
Smoo
th m
oder
ately
wea
ther
ed a
nd al
tere
d su
rface
s
Poor
Slic
kens
ided
hig
hly
wea
ther
ed su
rface
s with
com
pact
co
atin
gs o
r filli
ngs o
f ang
ular
frag
men
ts
Very
poo
rSl
icke
nsid
ed h
ighl
y w
eath
ered
surfa
ces w
ith so
cla
yco
atin
gs o
r infi
lling
s
80
70
60
50
40
30
20
10
RQD
2
Figure 9 GSI showing scale A represented by 15 JCond89 and scale B by RQD2 [13]
12 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
equations (22) and (23) [80 81] respectively when intactrock and rock mass constants (mb s and a) are known
Cm σci (1 + 2a)s + (1 minus a)mbσ3n1113858 1113859 mbσ3n + s( 1113857
aminus 1
(1 + a)(2 + a)
1 + 6amb mbσ3n + s( 1113857aminus 1(1 + a)(2 + a)1113872 1113873
1113969 (22)
φm sinminus 1 3amb mbσ3 n + s( 1113857aminus 1
(1 + a)(2 + a) + 3amb mbσ3n + s( 1113857aminus 1
⎡⎣ ⎤⎦ (23)
When the GSI value is known then equations (14)ndash(21)can be used to find the values ofmb s and a for estimation ofcohesion and angle of internal friction
52 Estimation of the Rock Mass Deformation ModulusRockmass deformationmodulus is an integral parameter fornumerical modeling of rock engineering problems for preand post failure analysis and assessment of the effectivenessof design [7] +e determination of rock mass deformationmodulus in the field is a very challenging job and the ob-tained results may be questionable due to complicatedprocedures [7 38 59 61 73 82] +erefore most of theresearchers prefer to estimate the rock mass deformationmodulus rather than determining in the field through in situtechniques [49 59 69ndash73 77 82ndash85] +e empirical modelsdeveloped based on GSI values for rock mass deformationmodulus are presented in Table 2
53 Estimation of the RockMass Strength and Poissonrsquos Ratio+e strength and passion ratio are very important param-eters used in numerical modeling of rock engineeringproblems for evaluating the behavior of rock mass +e insitu determination of these properties of rock mass usingdifferent methods is difficult and time-consuming as com-pared to empirical methods [7] +e following are differentempirical models using GSI and other parameters for es-timation of strength and Poissonrsquos ratio of a rock mass
Ramamurthy in 1993 established an empirical modelusing GSI and σc (compressive strength of the rock) forestimating the compressive strength of rock mass as given inthe following equation [91]
σc m σce(GSIminus 100)185
(35)
Vasarhelyi in 2009 [93] developed linear empiricalmodels as shown in equations (36) and (37) for estimationof Poissonrsquos ratio using GSI Poissonrsquos ratio for intact rock(vi) and HoekndashBrown constant (mi) [93]
vrm minus 0002GSI + vi + 02 (36)
vrm minus 0002GSI minus 0003mi + 0457 (37)
where vrm is Poissonrsquos ratio for rock mass and vi is intact-rock Poissonrsquos ratio
54 Relation of GSI with Other Rock Mass ClassificationSystems Hoek and Diederichs developed an alternateequation for estimation of the GSI using surface jointcharacteristicsratio as shown in the following equation[46]
GSI 15JCOND89 +RQD2
(38)
Equation (38) can be rearranged to estimate the value ofRQD as shown in the following equation
RQD 2(GSI minus 15JCOND89) (39)
Barton et al in 1974 suggested a relation representingthe detail of joint condition (joint alteration (Ja) jointroughness (Jr)) as given in the following equation
JCOND89 35JrJa
1 +(JrJa)1113888 1113889 (40)
Substituting equation (40) into equation (39) we will get
RQD 2(GSI minus 525)JrJa
1 +(JrJa)1113888 1113889 (41)
Various researchers worldwide have developed differ-ent correlations for estimating the GSI value using RMR76RMR89 Q value RMR RCR RSR and RMI as shown inTable 3
55Applicationof theGSI System inStabilityAnalysis of SlopeTunnels andExcavations Sonmez and Ulusay in 1999 [52]developed a refine quantitative numerical basis for esti-mating the GSI value as shown in Figure 5 +ey useddisturbance factor due to the method of excavation (mode ofblasting) and rock mass strength for estimating the GSIvalue For stability of slopes they used the GSI value as theinput in HOBRSLP software for slope stability analysis ofcircular and noncircular slip surfaces In 2013 Hoek et al[13] presented a simplified qualitative approach for stabilityanalysis of tunnels having s span of 10m and slopes having aheight less than 20m For larger caverns and slopes considerreducing value of the GSI for decreasing block interlockingFurthermore the GSI value is usually used as an inputparameter in RocLab software for evaluating differentstrength parameters of rock mass and numerical tools suchas RS2 and RS3 developed by Rocscience for numerical
Advances in Civil Engineering 13
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
modeling stability analysis of engineering structures andpre- and postbehavior of rock mass (tunnels slopes andexcavations) +e value of RMR Q-system and other em-pirical methods is not used as compared to the GSI valuedirectly in numerical tools but indirectly the supportsystem recommended by these empirical methods is in-stalled and evaluated in numerical modeling
56 Application of the GSI in Excavatability Assessment ofRock Masses GSI system has been successfully utilized inexcavability assessment of rock mass during construction ofunderground excavation +e mode of excavation is definedbased on 61 site investigations including sedimentary and
metamorphic rocks having deformed structures and jointconditions utilizing the GSI value by Tsiambaos and Sar-oglou [98] +e research study concluded that (1) when GSIvaluegt 65 and point load index (Is50)gt 3MPa and GSIvaluegt 60 and Is50lt 3MPa blasting is preferred to be usedin massive blocky having tight joint conditions (2) Whenthe GSI value ranges from 20 to 45 and Is50ge 3MPa and theGSI value ranges from 25 to 55 and Is50lt 3MPa excavationis carried out by ripping while in the transition phase fromblasting to ripping the hyudraulic breaker is to be used forexcavation +is excavability assessment is suitable to beused in rock mass where discontinuities control the exca-vation while it is not suitable in heterogeneous rock massessuch as sheared flysch bimrocks and soft rocks
Table 3 Comparison of various correlations among the rock mass classifications
S no Authors Correlation Estimated parameter Reference Equation no1 Hoek and Brown GSI RMR76 RMR76 gt 18 GSI from RMR79 [61] (42)2 Hoek and Brown GSI RMR89 minus 5 RMR89 gt 23 GSI from RMR89 [61] (43)3 Morales et al GSI 4714 + 0687RMR GSI from RMR89 [94] (44)
4 Cosar GSI 042RMR + 2307 GSI from RMR89 [95] (45)GSI 161LnQ + 4299 GSI from Q [95] (46)
5 Osgoui and Unal GSI 6e005RMR GSI from RMR89 [51] (47)
6 Hashemi et al GSI(Hoek1995) 0692RMR89 + 2232 GSI from RMR89 [84] (48)GSI(Hoek1995) 0917GSI(Cai2004) + 318 GSI from GSI [84] (49)
7 Irvani et al GSI 135RMR minus 164 GSI from RMR89 [96] (50)8 Singh and Tamrakar GSI 073RMR minus 438 GSI from RMR89 [65] (51)9 Zhang et al GSI 121RMR minus 1861 GSI from RMR89 [50] (52)
10 Sadeghi et al
GSI 05939RMR11047Basic GSI from RMRBasic
[97]
(53)GSI 09143RMR + 6132 GSI from RMR89 (54)GSI 12638LnQ + 28538 GSI from Q (55)
GSI 10951LnRMi + 33157 GSI from RMi (56)GSI 29891LnRSR minus 71285 GSI from RSR (57)
GSI 17RCR08566 GSI from RCR (58)GSI 25577Q02841
N GSI from QN (59)GSI 07861GSICai + 85483 GSI from GSICai (60)
Table 2 Empirical models for rock mass deformation modulus
Sno
Empiricalmodel Unit Name of the
researcher Parameters Year References Equationno
1 Erm Ei(sa)04Ei 50GPa s e((GSIminus 100)9)a 05 + (16)(e(minus GSI15) minus e(minus 203))
GPa Sonmez et al Ei GSI sand a 2004 [86] (24)
2 Erm (1 minus (D2) )(σc100)
111396810((GSIminus 10)40) for σc lt 100MPa GPa Hoek et al GSI D and
σc
2002 [78] (25)
3 Erm (1 minus (D2))10((GSIminus 10)40) for σc gt 100MPa GPa Hoek et al GSI D andσc
2002 [78] (26)
5 Erm Ei 002 + ((1 minus (D2))(1 + exp((60 + 15D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs
GSI D andEi 2006 [77] (27)
6 Erm 105 ((1 minus (D2))(1 + exp((75 + 25D minus GSI)11)))1113864 1113865 MPa Hoek andDiederichs GSI and D 2006 [77] (28)
7 Erm Ei(e(GSIminus 100)A) GPa Van andVasarhelyi
GSI Eiand A 2010 [87] (29)
8 Erm tan(
156 + (ln(GSI))21113969
)σc
3radic GPa Beiki et al GSI and σc 2010 [88] (30)
9 Erm (σc100)
1113968lowast 10(GSIminus 10)40 for σc lt 100MPa GPa Palmstrom
and Singh σc and GSI 1998 [89] (31)
10 Erm Ei(s)34 where s e((GSIminus 100)9) GPa Sonmez et al Ei GSIand s 2004 [59] (32)
11 Erm 035lowast exp006GSI GPa Majdi et al GSI 2012 [90] (33)12 Erm
(Ei100)
111396810((GSIminus 20)35) GPa Majdi et al GSI 2012 [90] (34)
14 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
6 Limitations
+e GSI system is based on assumptions that the rock massbehaves like isotropic media and the behavior of rock massis not reliant on the direction of the applied load Based onGSI assumptions this system should not be used in suchconditions where predominant structural orientation andstructurally dependent gravitational instability exist
+e GSI system is not applied to already excavated facesin very hard rock having few discontinuities spaced atdistances of similar magnitude to the dimensions of thetunnel or slope under consideration because in this case thestability of the tunnel or slope is controlled by the three-dimensional geometry of the intersecting discontinuities andthe free faces created by the excavation
+e quantitative version of the GSI developed by HSonmez and Ulusay is valid to use in such cases in whichfrequency and orientation of discontinuities are measuredeasily+e quantification version of the GSI is not effective intectonically disturbed rock masses having destroyed struc-tural fabric In such a case the researchers recommend usinga qualitative version of the GSI based on careful visualobservations
7 Conclusions
+e following conclusions are drawn from this study
(1) +e empirical methods such as RMR Q and RMIused RQD as input parameters which are notsuitable to be applied in a very weak and highlyjointed rock mass environment because for suchrock mass conditions RQD value is taken as zero
(2) +e empirical methods (RMR Q-system RMI etc)are inconvenient to be used for design of tunnelscaverns and other underground excavations insqueezing high-stress and weak rock massenvironments
(3) +e empirical methods (RMR Q-system RMI etc)are not suitable to be used for prepost failure andstability analysis of engineering projects throughnumerical modeling
(4) +e GSI system as compared to other rock massclassification systems may represent the hetero-geneity of rock mass in a convenient way in terms ofrock mass structure domain
(5) +ree different approaches are effectively used toestimate the GSI value ie qualitative quantitativeand empirical models based on other empiricalmethods +e qualitative and quantitative ap-proaches are suitable to be used for evaluating theweak jointed layered and heterogeneous rockmass for engineering design purpose
(6) It has been concluded from the study that the RMRQ-system and other empirical methods are con-venient to be used in favorable joint conditions orpresence of less joints in the hard rock mass en-vironment while the GSI system is recommended
to be used for weak to very weak and high-stressrock mass environments
(7) +e GSI system is explicitly used for estimation ofgeomechanical properties of rock mass with max-imum confidence level as compared to other rockmass classification systems
(8) GSI has a wide range of applications in rock en-gineering However this system did not cover thedesign of the support system for tunnel and otherstructures as compared to RMR and Q-system
(9) +e qualitative approach for estimating the GSIsystem required better knowledge and skills aboutcollecting and evaluating the field data Howeverthe quantitative and empirical model approachesare easy to be used for the assessment of rock massenvironment
(10) +is study will provide a common understanding tofield professionals to use different estimating ap-proaches for the GSI value in a convenient way forevaluating the rock mass environment and esti-mating the required rock mass strength and de-formation properties for numerical modeling
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Supplementary Materials
Figure 1 GSI charts for confined and fissile molaissic rockmasses for tunnels and surface excavations respectively [64]Hoek et al in 2005 [64] incorporated weak heterogeneousrock masses and lithological variability of a rock mass Hoeket al developed two charts for confined and fissile molaissicrock masses for tunnels and surface excavations as pre-sented in Figures 1(a) and 1(b) respectively Figure 2 scale Aand B inserted at x- and y-axis of the GSI chart [13](Supplementary Materials)
References
[1] L Jing ldquoA review of techniques advances and outstandingissues in numerical modelling for rock mechanics and rockengineeringrdquo International Journal of Rock Mechanics andMining Sciences vol 40 no 3 pp 283ndash353 2003
[2] M N Bidgoli Z Zhao and L Jing ldquoNumerical evaluation ofstrength and deformability of fractured rocksrdquo Journal of RockMechanics and Geotechnical Engineering vol 5 no 6pp 419ndash430 2013
[3] L Jing and J A Hudson ldquoNumerical methods in rock me-chanicsrdquo International Journal of Rock Mechanics and MiningSciences vol 39 no 4 pp 409ndash427 2002
[4] J A Hudson and X-T Feng ldquoTechnical auditing of rockmechanics modelling and rock engineering designrdquo
Advances in Civil Engineering 15
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
International Journal of Rock Mechanics and Mining Sciencesvol 47 no 6 pp 877ndash886 2010
[5] T Dalgleish Engineering RockMechanics vol 136 no 1 2007[6] M Pinheiro X Emery T Miranda L Lamas and M Espada
ldquoModelling geotechnical heterogeneities using geostatisticalsimulation and finite differences analysisrdquo Minerals vol 8no 2 pp 52ndash19 2018
[7] S Hussain M Khan Z Ur Rahman et al ldquoEvaluating thepredicting performance of indirect methods for estimation ofrock mass deformation modulus using inductive modellingtechniquesrdquo Journal of Himalayan Earth Sciences vol 51no 1 pp 61ndash74 2018
[8] S Hussain N Mohammad M Tahir Z Ur Rehman andN Mohammad ldquoRock mass characterization along the tunnelaxis for Golen Gol hydropower project Chitral PakistanrdquoJournal of Himalayan Earth Sciences vol 49 no 2 pp 75ndash832016
[9] Z Ur Rehman N Mohammad S Hussain and M TahirldquoNumerical modeling for the engineering analysis of rockmass behaviour due to sequential enlargement of Lowaritunnel Chitral Khyber Pakhtunkhwa Pakistanrdquo InternationalJournal of Geotechnical Engineering vol 13 no 1 pp 1ndash72019
[10] S Hussain Z Ur Rehman N Mohammad et al ldquoNumericalmodeling for engineering analysis and designing of optimumsupport systems for headrace tunnelrdquo Advances in CivilEngineering vol 2018 Article ID 7159873 10 pages 2018
[11] E Hoek and E T Brown ldquo+e Hoek-Brown failure criterionand GSImdash2018 editionrdquo Journal of Rock Mechanics andGeotechnical Engineering vol 11 no 3 pp 445ndash463 2019
[12] G F Andriani and M Parise ldquoApplying rock mass classifi-cations to carbonate rocks for engineering purposes with anew approach using the rock engineering systemrdquo Journal ofRock Mechanics and Geotechnical Engineering vol 9 no 2pp 364ndash369 2017
[13] E Hoek T G Carter andM S Diederichs ldquoQuantification ofthe geological strength index chartrdquo in Proceedings of the 47thUS Rock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[14] V Marinos and T G Carter ldquoMaintaining geological realityin application of GSI for design of engineering structures inrockrdquo Engineering Geology vol 239 pp 282ndash297 2018
[15] K Terzaghi Rock Defects and Loads on Tunnel SupportsHarvard University Cambridge MA USA 1946
[16] H Lauffer ldquoGebirgsklassifizierung fur den Stol LenbaurdquoGeology Bauwesen vol 24 pp 46ndash51 1958
[17] D U Deere ldquoTechnical description of rock cores for engi-neering purposesrdquo Rock Mechanics and Engineering GeologySpringer Berlin Germany 1963
[18] D U Deere and R P Miller Engineering Classification andIndex Properties for Intact Rock National Technical Infor-mation Service Springfield VA USA 1966
[19] G EWickham H R Tiedemann and E H Skinner ldquoSupportdeterminations based on geologic predictionsrdquo in Proceedingsof the North American Rapid Excavation and TunnelingConference pp 5ndash7 Chicago IL USA 1972
[20] Z T Bieniawski Engineering Rock Mass Classifications AComplete Manual for Engineers and Geologists in MiningCivil and Petroleum Engineering Wiley Hoboken NJ USA1989
[21] Z T Bieniawski Engineering Classification of Jointed RockMasses South African Institute of Civil Engineers MidrandSouth Africa 1973
[22] N Barton R Lien and J Lunde ldquoEngineering classification ofrock masses for the design of tunnel supportrdquo Rock Me-chanics vol 6 pp 189ndash236 1974
[23] G J Pacher and L Rabcewicz ldquoZum der seitigen stand dergebirgs e klassifizierung in stollen-und tunnelbaurdquo in Pro-ceedings of the 22nd Geomechanics Colloquia p 51 1974
[24] J A Franklin ldquoSafety and economy in tunnelingrdquo in Pro-ceedings of the 10th Canadian Rock Mechanics Symposiumpp 27ndash53 Kingston Canada 1975
[25] ISRM ldquoCommission on classification of rocks and rockmasses basic geotechnical description of rock massesrdquo In-ternational Journal of Rock Mechanics and Mining Science ampGeomechanics Abstracts vol 2 pp 85ndash110 1981
[26] A F H Stille and T Groth FEM-analysis of Rock MechanicsProblems by JOBFEM Swedish Rock Engineering ResearchFoundation Publication Stockholm Sweden 1982
[27] D A Williamson and C R Kuhn ldquo+e unified rock clas-sification systemrdquo in Rock Classification Systems for Engi-neering Purposes ASTM West Conshohocken PA USA1984
[28] A Palmstrom RMimdasha Rock Mass Characterization System forRock Engineering Purposes University of Oslo Oslo Norway1995
[29] E Hoek and P K Kaiser Support of Underground Excavationsin Hard Rock CRC Press Boca Raton FL USA 1997
[30] N Barton ldquoTBM perfomance estimation in rock usingQTBMrdquo Tunnel and Tunnelling International pp 30ndash341999
[31] Z Sen and B H Bahaaeldin ldquoModified rock mass classifi-cation system by continuous ratingrdquo Engineering Geologyvol 67 pp 269ndash280 2003
[32] Z T Bieniawski von Prein B C Tamames J M GaleraFernandez and M Hernandez Alvarez ldquoRock mass excav-ability indicator new way to selecting the optimum tunnelconstruction methodrdquo Tunnelling and Underground SpaceTechnology vol 3 p 237 2006
[33] O Aydan R Ulusay and N Tokashiki ldquoRock mass qualityrating (RMQR) system and its application to the estimation ofgeomechanical characteristics of rock massesrdquo in EngineeringGeology for Society and TerritorymdashVolume 6 Applied Geologyfor Major Engineering Projects Springer Berlin Germany2015
[34] M Mohammadi and M F Hossaini ldquoModification of rockmass rating system interbedding of strong and weak rocklayersrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 6 pp 1165ndash1170 2017
[35] C Saroglou S Qi S Guo and F Wu ldquoARMR a newclassification system for the rating of anisotropic rockmassesrdquo Bulletin of Engineering Geology and the Environmentvol 78 no 5 pp 3611ndash3626 2019
[36] H Rehman W Ali A Naji J-j Kim R Abdullah andH-k Yoo ldquoReview of rock-mass rating and tunneling qualityindex systems for tunnel design development refinementapplication and limitationrdquo Applied Sciences vol 8 no 8p 1250 2018
[37] X Wang J Lai R S Garnes and Y Luo ldquoSupport system fortunnelling in squeezing ground of Qingling-Daba moun-tainous area a case study from soft rock tunnelsrdquo Advances inCivil Engineering vol 2019 Article ID 8682535 17 pages2019
[38] M Romana J B Seron and EMontalar ldquoSMR geomechanicsclassification application experience and validationrdquo inProceedings of the 10th ISRM Congress Sandton South AfricaSeptember 2003
16 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
[39] R Bertuzzi ldquoRevisiting rock classification to estimate rockmass propertiesrdquo Journal of Rock Mechanics and GeotechnicalEngineering vol 11 no 3 pp 494ndash510 2019
[40] P Marinos V Marinos and E HoekGe Geological StrengthIndex (GSI) A Characterization Tool for Assessing EngineeringProperties of Rock Masses Taylor and Francis Abingdon UK2007
[41] V Marinos P Marinos and E Hoek ldquo+e geological strengthindex applications and limitationsrdquo Bulletin of EngineeringGeology and the Environment vol 64 no 1 pp 55ndash65 2005
[42] P Marinos E Hoek and V Marinos ldquoVariability of theengineering properties of rock masses quantified by thegeological strength index the case of ophiolites with specialemphasis on tunnellingrdquo Bulletin of Engineering Geology andthe Environment vol 65 pp 129ndash142 2006
[43] E Hoek P Marinos and M Benissi ldquoApplicability of thegeological strength index (GSI) classification for very weakand sheared rock masses +e case of the Athens Schistformationrdquo Bulletin of Engineering Geology and the Envi-ronment vol 57 no 2 pp 151ndash160 1998
[44] E Hoek D Wood and S Shah ldquoA modified Hoek-Browncriterion for jointed rock massesrdquo in Proceedings of the RockMechanic Symposium pp 209ndash214 Chester UK 1992
[45] P Marinos and E Hoek ldquoEstimating the geotechnicalproperties of heterogeneous rock masses such as FlyschrdquoBulletin of Engineering Geology and the Environment vol 60no 2 pp 85ndash92 2001
[46] E Hoek and M S Diederichs ldquoQuantification of the geo-logical strength index chartrdquo in Proceedings of the 2013 USRock MechanicsGeomechanics Symposium San FranciscoCA USA 2013
[47] H Sonmez and R Ulusay ldquoA discussion on the Hoek-Brownfailure criterion and suggested modifications to the criterionverified by slope stability case studiesrdquo Bulletin of EarthSciences vol 26 pp 77ndash99 2002
[48] P Marinos and E Hoek ldquoGSI a geologically friendly tool forrock mass strength estimationrdquo in Proceedings of the 2000International Conference on Geotechnical and GeologicalEngineering Melbourne Australia 2000
[49] M Cai P K Kaiser H Uno Y Tasaka and M MinamildquoEstimation of rock mass deformation modulus and strengthof jointed hard rock masses using the GSI systemrdquo Inter-national Journal of Rock Mechanics and Mining Sciencesvol 41 no 1 pp 3ndash19 2004
[50] Q Zhang X Huang H Zhu and J Li ldquoQuantitative as-sessments of the correlations between rockmass rating (RMR)and geological strength index (GSI)rdquo Tunnelling and Un-derground Space Technology vol 83 pp 73ndash81 2019
[51] R R Osgoui and E Unal ldquoAn empirical method for design ofgrouted bolts in rock tunnels based on the geological strengthindex (GSI)rdquo Engineering Geology vol 107 no 3-4pp 154ndash166 2009
[52] H Sonmez and R Ulusay ldquoModifications to the geologicalstrength index (GSI) and their applicability to stability ofslopesrdquo International Journal of Rock Mechanics and MiningSciences vol 36 pp 743ndash760 1999
[53] K Hong E Han and K Kang ldquoDetermination of geologicalstrength index of jointed rock mass based on image pro-cessingrdquo Journal of Rock Mechanics and Geotechnical Engi-neering vol 9 no 4 pp 702ndash708 2017
[54] H Sonmez C Gokceoglu and R Ulusay ldquoAn application offuzzy sets to the geological strength index (GSI) system usedin rock engineeringrdquo Engineering Applications of ArtificialIntelligence vol 16 no 3 pp 251ndash269 2003
[55] G Somodi A Krupa L Kovacs and B Vasarhelyi ldquoCom-parison of different calculation methods of geological strengthindex (GSI) in a specific underground construction siterdquoEngineering Geology vol 243 pp 50ndash58 2018
[56] R R Osgoui R Ulusay and E Unal ldquoAn assistant tool for theGeological Strength Index to better characterize poor and verypoor rock massesrdquo International Journal of Rock Mechanicsand Mining Sciences vol 47 no 4 pp 690ndash697 2010
[57] N Deisman M Khajeh and R J Chalaturnyk ldquoUsinggeological strength index (GSI) to model uncertainty in rockmass properties of coal for CBMECBM reservoir geo-mechanicsrdquo International Journal of Coal Geology vol 112pp 76ndash86 2013
[58] Z Gurocak P Solanki and M M Zaman ldquoEmpirical andnumerical analyses of support requirements for a diversiontunnel at the Boztepe dam site eastern Turkeyrdquo EngineeringGeology vol 91 no 2ndash4 pp 194ndash208 2007
[59] H Sonmez C Gokceoglu H A Nefeslioglu and A KayabasildquoEstimation of rock modulus for intact rocks with an artificialneural network and for rock masses with a new empiricalequationrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 224ndash235 2006
[60] Z T Biniawski ldquoClassification of rockmasses for engineeringthe RMR system and future trendsrdquo in Rock Testing and SiteCharacterization pp 553ndash573 Elsevier Amsterdam Neth-erlands 1993
[61] E Hoek and E T Brown ldquoPractical estimates of rock massstrengthrdquo International Journal of Rock Mechanics andMining Sciences vol 34 no 8 pp 1165ndash1186 1997
[62] D U Deere Rock Quality Designation (RQD) after 20 YearsEngineer Waterways Library Vicksburg MS USA 1989
[63] E Hoek ldquoStrength of rock and rock massesrdquo ISRM NewsJournal vol 2 pp 4ndash16 1994
[64] E Hoek P G Marinos and V P Marinos ldquoCharacterisationand engineering properties of tectonically undisturbed butlithologically varied sedimentary rock massesrdquo InternationalJournal of Rock Mechanics and Mining Sciences vol 42 no 2pp 277ndash285 2005
[65] J L Singh and N K Tamrakar ldquoRock mass rating andgeological strength index of rock masses of +opal-Malekhuriver areas central Nepal lesser Himalayardquo Bulletin of theDepartment of Geology vol 16 pp 29ndash42 2013
[66] M Rasouli ldquoEngineering geological studies of the diversiontunnel focusing on stabilization analysis and support designIranrdquo Engineering Geology vol 108 no 3-4 pp 208ndash2242009
[67] S K Dwivedi P C Adhikary and J M a N TamrakarEngineering Geological and Geotechnical Characteristics of theKankai Hydro-Power Tunnel in Soft Rock Nepal IAEGVienna Austria 2006
[68] M Genis H Basarir A Ozarslan E Bilir and E BalabanldquoEngineering geological appraisal of the rock masses andpreliminary support design Dorukhan tunnel ZonguldakTurkeyrdquo Engineering Geology vol 92 no 1-2 pp 14ndash26 2007
[69] C O Aksoy M Genis G Uyar Aldas V Ozacar S C Ozerand O Yılmaz ldquoA comparative study of the determination ofrock mass deformation modulus by using different empiricalapproachesrdquo Engineering Geology vol 131-132 pp 19ndash282012
[70] V K Singh D Singh and T N Singh ldquoPrediction of strengthproperties of some schistose rocks from petrographic prop-erties using artificial neural networksrdquo International Journalof Rock Mechanics and Mining Sciences vol 38 no 2pp 269ndash284 2001
Advances in Civil Engineering 17
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
[71] J Shen M Karakus and C Xu ldquoA comparative study forempirical equations in estimating deformation modulus ofrock massesrdquo Tunnelling and Underground Space Technologyvol 32 pp 245ndash250 2012
[72] A Kayabasi C Gokceoglu andM Ercanoglu ldquoEstimating thedeformation modulus of rock masses a comparative studyrdquoInternational Journal of Rock Mechanics and Mining Sciencesvol 40 no 1 pp 55ndash63 2003
[73] J Gholamnejad H R Bahaaddini and M Rastegar ldquoPre-diction of the deformation modulus of rock masses usingartificial neural networks and regression methodsrdquo Journal ofMining amp Environment vol 4 no 1 pp 35ndash43 2013
[74] E Eberhardt ldquo+e Hoek-Brown failure criterionrdquo RockMechanics and Rock Engineering vol 45 no 6 pp 981ndash9882012
[75] E Hoek and E T Brown ldquoHoek-Brown failure criterionmdasha1988 updaterdquo International Journal of Rock Mechanics andMining Sciences amp Geomechanics Abstracts 1990
[76] E Hoek C Carranza and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 2002 Narms-Tac Conference Toronto Canada 2002
[77] E Hoek and M S Diederichs ldquoEmpirical estimation of rockmass modulusrdquo International Journal of Rock Mechanics andMining Sciences vol 43 no 2 pp 203ndash215 2006
[78] E Hoek C C Torres and B Corkum ldquoHoek-Brown failurecriterionmdash2002 editionrdquo in Proceedings of the 5th NorthAmerican Rock Mechanics Symposium Toronto Canada2002
[79] E Hoek and E T Brown Underground Excavation in RockInstitution of Mining and Metallurgy London UK 1980
[80] Y C Yang X G Yang and H Ge Xing ldquoEstimation methodfor shear strength parameters of fault rock based on geologicalstrength indexrdquo Electronic Journal of Geotechnical Engi-neering vol 21 no 5 pp 1715ndash1725 2016
[81] K-S Kang N-L Hu C-S Sin S-H Rim E-C Han andC-N Kim ldquoDetermination of the mechanical parameters ofrock mass based on a GSI system and displacement backanalysisrdquo Journal of Geophysics and Engineering vol 14 no 4pp 939ndash948 2017
[82] HMohammadi and R Rahmannejad ldquo+e estimation of rockmass deformation modulus using regression and artificialneural notworks analysisrdquo Arabian Journal for Science andEngineering vol 35 no 1 pp 205ndash217 2010
[83] B Figueiredo L Lamas and J Muralha ldquoDetermination of insitu stresses using large flat jack testsrdquo in Proceedings of theISRM International Symposium 2010 New Delhi India 2010
[84] M Hashemi AElig S Moghaddas and R Ajalloeian ldquoAppli-cation of rock mass characterization for determining themechanical properties of rock mass a comparative studyrdquoRock Mechanics and Rock Engineering vol 43 pp 305ndash3202010
[85] E Ghotbi Ravandi R Rahmannejad A E Feili Monfaredand E Ghotbi Ravandi ldquoApplication of numerical modelingand genetic programming to estimate rock mass modulus ofdeformationrdquo International Journal of Mining Science andTechnology vol 23 no 5 pp 733ndash737 2013
[86] H Sonmez C Gokceoglu and R Ulusay ldquoIndirect deter-mination of the modulus of deformation of rock masses basedon the GSI systemrdquo International Journal of Rock Mechanicsand Mining Sciences vol 41 no 5 pp 849ndash857 2004
[87] P Van and B Vasarhelyi ldquoRelation of rock mass charac-terization and damagerdquo Rock Engineering in Difficult GroundConditions (Soft Rocks and Karst) pp 399ndash404 CRC PressBoca Raton FL USA 2010
[88] M Beiki A Bashari and A Majdi ldquoGenetic programmingapproach for estimating the deformation modulus of rockmass using sensitivity analysis by neural networkrdquo Interna-tional Journal of Rock Mechanics and Mining Sciences vol 47no 7 pp 1091ndash1103 2010
[89] A Palmstrom and R Singh ldquo+e deformation modulus ofrock massesmdashcomparisons between in situ tests and indirectestimatesrdquo Tunnelling and Underground Space Technologyvol 16 no 2 pp 115ndash131 2001
[90] A Majdi A Bashari and M Beiki ldquoEstimation of rock massdeformation modulus based on GSI systemrdquo HarmonisingRock Engineering and the Environment pp 2113ndash2117 2012
[91] T Ramamurthy ldquoStrength and modulus responses of an-isotropic rocksrdquo in Comprehensive Rock Engineering Prin-ciple Practice amp Projects pp 313ndash329 Elsevier AmsterdamNetherlands 1993
[92] B Vasarhelyi ldquoA possible method for estimating the Poissonrsquosrate values of the rock massesrdquo Acta Geodaetica et GeophysicaHungarica vol 44 no 3 pp 313ndash322 2009
[93] E Osgoui and R Unal ldquoRock reinforcement design for un-stable tunnels originally excavated in very poor rock massUnderground space use in analysis of the past and lessons forthe futurerdquo in Proceedings of the International World TunnelCongress and the 31st ITA General Assembly pp 291ndash296Istanbul Turkey 2005
[94] T Morales G Uribe-Etxebarria and J A Uriarte ldquoGeo-mechanical characterisation of rock masses in Alpine regionsthe Basque arc (Basque-Cantabrian basin northern Spain)rdquoEngineering Geology vol 71 no 3-4 pp 343ndash362 2004
[95] S Cosar Application of rock mass classification systems forfuture support design of the Dim tunnel near Alanya PhDthesis Middle East Technical University Ankara Turkey2004
[96] I Irvani W Wilopo and D Karnawati ldquoDetermination ofnuclear power plant site in west Bangka based on rock massrating and geological strength indexrdquo Journal of AppliedGeology vol 5 no 2 2015
[97] S Sadeghi E S Teshnizi and B Ghoreishi ldquoCorrelationsbetween various rock mass classificationcharacterizationsystems for the Zagros tunnel-W Iranrdquo Journal of MountainScience vol 17 no 7 2020
[98] G Tsiambaos and H Saroglou ldquoExcavatability assessment ofrock masses using the geological strength index (GSI)rdquoBulletin of Engineering Geology and the Environment vol 69no 1 pp 13ndash27 2010
18 Advances in Civil Engineering
![ReviewArticle - Hindawi Publishing Corporationdownloads.hindawi.com/journals/crp/2020/4150291.pdf2019/12/20 · postmenopausalwomen[8–10],consistentwiththeobser-vationthatwomenolderthan55yearshavea10-foldgreater](https://static.fdocuments.in/doc/165x107/609acb72d1e81e0b02201a33/reviewarticle-hindawi-publishing-20191220-postmenopausalwomen8a10consistentwiththeobser-vationthatwomenolderthan55yearshavea10-foldgreater.jpg)
