Seismic Performance Evaluation of Reinforced Plaster...

www.cafetinnova.org

Indexed in

Scopus Compendex and Geobase Elsevier, Chemical

Abstract Services-USA, Geo-Ref Information Services-USA

ISSN 0974-5904, Volume 05, No. 02

April 2012, P.P. 193-205

#02050201 Copyright © 2012 CAFET-INNOVA TECHNICAL SOCIETY. All rights reserved.

Seismic Performance Evaluation of Reinforced Plaster Retrofitting

Technique for Low-Rise Block Masonry Structures

NAVEED AHMAD1, QAISAR ALI

2, MOHAMMAD ASHRAF

3, AKHTAR NAEEM

3 and BASHIR ALAM

3

1ROSE School-IUSS Pavia, Pavia, Italy.

2Earthquake Engineering Center, University of Engineering & Technology (UET), Peshawar, Pakistan.

3Department of Civil Engineering, University of Engineering & Technology (UET), Peshawar, Pakistan.

Email: [email protected]

Abstract: Low-rise unreinforced block masonry structures are investigated for resistance against site amplified

ground motions, before and after retrofitting with reinforced plaster. The aim of the study was to evaluate the

effectiveness of indigenous and cost-effective retrofitting technique for improving the seismic performance of

ordinary masonry structures. The retrofitting technique involved cement grouting of damaged walls and fixing steel

wire mesh on wall surface, plastered with rich cement mortar. Experimental investigations carried out on block

masonry material and prototype walls are analyzed to retrieve the mechanical characteristics for mathematical

modeling of structures. Two case study low-rise (two storey) structures are assessed through nonlinear incremental

dynamic time history analysis technique using natural accelerograms. The considered retrofitting technique

increases the lateral load carrying capacity of in-plane walls by about 10 percent and ductility capacity by 100

percent and ensures flexure (toe crushing) response mechanism to lateral loading. The technique increased the

seismic resistance of structures to ground shaking by 43 percent. It is observed that the retrofitted structures can

survive ground shaking intensity up to 0.40g/0.50g whereas shaking intensity of 0.73g/0.90g (and above) may cause

the structure collapse (with 95 percent confidence), indicating that the technique can be confidently used for

improving the seismic performance of block masonry structures.

Keywords: block masonry structures; reinforced plaster; ferrocement; cost-effective retrofitting, structure useful

life.

1. Introduction:

Significant proportion of building stock in urban areas

are constructed using concrete block masonry in

cement-mortar. For example, concrete blocks are used

in construction of walls in the order of 10-20 percent in

most of the Districts of Punjab and Sindh Provinces of

Pakistan. In few other cases the percentage can reach

event up to 40-60 percent e.g. District Karachi in Sindh

and few other Districts in Northern areas of Khyber

Pakhtoonkhwa of Pakistan (ERRA, 2006; Maqsood and

Schwarz, 2008). Concrete block masonry is used mostly

for one to two storey residential structures in urban

areas. However, concrete block masonry walls are also

used as an infill panel in reinforced concrete frame

construction e.g. the City like Karachi (Badrashi et al.,

2010) among others. Also, recently following major

earthquake in Pakistan, the use of concrete block units

has been increased for the re-construction activity in

major affected areas (both urban and rural) due to the

fact that concrete blocks can be easily and rapidly

produced using even small machines (Stephenson,

2008).

In the recent major earthquake Mw 7.6 2005 Kashmir,

about 30 percent of the building stock subjected to

ground shaking included block masonry structures in

the nearby urban areas. Almost 60 percent of block

masonry structures were either heavily damaged are

collapsed (Naseer et al., 2010; Rossetto and Peiris,

2009), which is significant. The recent investigation on

the seismic risk assessment of existing structures in

Pakistan reported that regional block masonry structures

subjected to ground shaking intensity of 0.30g can result

in the collapse of 70 percent of the exposed building

stock (Ahmad, 2011), which is due to the extremely low

mechanical characteristics of existing block masonry

material. This means that existing block masonry

construction is at risk in the high hazard zones of the

country e.g. Zone 3 (0.24g–0.32g) and Zone 4 (> 0.32g)

as specified by BCP (2007), indicating that ordinary

block masonry structures cannot be confidently

constructed in high hazard zones. Similar, findings have

been reported also by Mahmood and Ingham (2011)

however observing also that significant amount of block

masonry typologies are earthquake-prone even in

moderate hazard zones e.g. Zone 2B (0.16–0.24g). This

situation motivate research on the seismic vulnerability

194 NAVEED AHMAD, QAISAR ALI, MOHAMMAD ASHRAF,

AKHTAR NAEEM and BASHIR ALAM

International Journal of Earth Sciences and Engineering

ISSN 0974-5904, Vol. 05, No. 02, April 2012, pp. 193-205

assessment of individual block masonry structures

toward disaster risk reduction (saving life during strong

ground motions) using efficient, yet affordable,

retrofitting techniques.

Cost-effective and efficient retrofitting techniques are

essential, particularly for the developing part of the

world exposed to moderate to strong ground motions.

Retrofitting techniques for masonry structures are

evolved to a much mature stage (ElGawady et al.,

2004), however its design and application may still be

affected by many factors like the structures typology

and its existing condition, performance objective, cost

of retrofitting, available time, resources and materials,

preservation of historical value etc., particularly in the

developing part of the world e.g. Pakistan, India, Iran,

among others which predominantly include masonry

structures. This clearly indicates the importance of

research on the investigation of efficient and cost-

effective retrofitting techniques for masonry structures.

The paper thus presents investigation carried out on

low-rise block masonry structures practiced mostly in

the urban areas of Pakistan, for the seismic resistance

evaluation of existing structures and their possible

improvement using indigenous and cost-effective

retrofitting technique. It included seismic performance

evaluation of structures; existing and retrofitted

(damaged structures) with reinforced plaster technique

developed and practiced at the Earthquake Engineering

Center of Peshawar. Experimental investigations carried

out recently at the Earthquake Engineering Center and

University of Engineering and Technology (UET)

Peshawar (Shoaib et al., 2011) on block masonry

assemblages and prototype walls (before and after

retrofitting) are considered for design and mathematical

modelling of case study structures.

Nonlinear dynamic time history analyses of four case

study structures are performed using natural

accelerograms, where the structures are assessed using

probabilistic approach taking into account the variability

of ground motions in performance evaluation. Ground

motions with different exceedance probability is

obtained for specified performance level of structures.

Example structures are investigated for the seismic

performance evaluation of the retrofitting technique

considering a single scenario earthquake and possible

events over the design life.

Masonry Walls Retrofitting Technique (Reinforced

Plaster):

The retrofitting technique investigated in the present

study included grout injection of cracks (for already

damaged walls) and application of steel welded wire

mesh on the surface of walls, plastered with rich

cement-sand mortar, the so called ferrocement overlay:

a thin layer of reinforced concrete consists of closely

spaced small size wire-mesh completely embedded in

cement mortar (ACI 549R-97, 1997). The technique

require relatively low level technical skills for

fabrication and application of ferrocement. The

provision of closely spaced steel wire-mesh in the

ferrocement overlay provides significant tensile strength

to retrofitted walls with low economic cost, without

adding significant additional mass to the structural

system, which makes the technique highly appealing in

developing part of the world exposed to significant

earthquakes.

1.1 Reinforced Plaster Retrofit Design Scheme:

In the first phase a pre-fabricated wire mesh is fixed on

the wall surface through fixing screws, where the

screws are fastened in the plastic plugs inserted in the

wall. Holes are formed in masonry units for inserting

plugs. This technique is recently developed and

practiced at the Earthquake Engineering Center of

Peshawar. Figure 1 depicts the schematic presentation

of reinforced plastering of existing masonry walls. For

the case of already damaged walls, the grouting of

cracks is also needed to be performed before plastering

mesh-fixed wall.

Recently, Shoaib et al. (2011) investigated this

retrofitting technique experimentally for the

performance improvement of damaged block masonry

prototype walls. In this case, the grouting of cracks is

performed 24 hours after the plastering of mesh-fixed

wall. For this purpose, nozzles (10mm diameter and

80mm length) were pre-inserted in the cracks at a

distance of 300mm to 500mm along the cracks soon

after fixing the wire mesh, where grout is injected to the

cracks with pressure pump. The steps involved in

grouting technique were outlined previously by Calvi

and Magenes (1994), Hamid et al. (1994) and Schuller

et al. (1994). The grouting injection material included a

cement-based grout comprised of 10 parts of Portland

cement, one part of lime and ultra expansion agent

employed at a rate of 250 grams per 50 kilograms of

cement (in the proportion of 200:20:1) with a water

cement ratio of 0.8. The wire mesh was designed of a

12.70mm x 12.70mm galvanized steel mesh made of

1mm diameter wires having 200 Mpa tensile strength.

The mesh is fixed to the wall surface by 40mm long

No.10 screws and plastic Ravel plug, provided as two

plugs per 100 square centimeter, called as 18 gauge wire

mesh. The plaster consisted of cement-sand mortar

which is prepared in proportion of 1:4 and applied in

average thickness of 20mm.

195 Seismic Performance Evaluation of Reinforced Plaster Retrofitting Technique for

Low-Rise Block Masonry Structures



Figure 1: Schematic of Reinforced Plaster Retrofitting Technique for Existing Masonry Walls Developed and

Practiced at the Earthquake Engineering Center of Peshawar. From Left to Right and Top to Bottom: Required

Accessories, Drilling for Inserting Screw Plugs, Application and Fixing of Wire Mesh and Plastering of Wall

(Shahzada, 2006)

1.2 Observed Behavior of Block Masonry Walls

(Before and After Retrofitting):

This section is largely based on the experimental study

carried out by Shoaib et al. (2011) on block masonry

material and walls and the reinforced plaster retrofitting

technique recently developed and practiced by the

Earthquake Engineering Center of Peshawar. The

following sections briefly describe the testing program,

the mechanical characteristics of block masonry and in-

plane response of case study block masonry walls

(before and after retrofitting) important within the scope

of present research study.

1.2.1 Experimental Program:

The experimental investigation included in-plane quasi-

static cyclic test on six masonry walls: three sample

walls were representative of a short pier in single storey

structures (and first floor of double storey structures),

other three sample walls were representative of a ground

floor short pier in double storey structures. The piers

were first tested in the unreinforced state which were

retrofitted and tested again. The tested walls were

deformed into life safety performance range till the peak

load was developed, strength and stiffness degradation

is noticed and when repair to damage was feasible.

Table 1: Material Properties of Block Masonry Walls

Tested at UET Peshawar, after Shoaib et al. (2011)

Property Observation

Block (Solid) Dimensions 150mmx200mmx300mm

Mortar Composition 1:6 (Cement-Sand)

Block Compressive

Strength 3.97 MPa

Mortar Compressive

Strength 3.60 MPa

Masonry Compressive

Strength 2.23 MPa

Diagonal Tensile Strength 0.20 MPa

Water Absorption 8.60%

Block Modulus of

Rupture 1.93 MPa

Friction Coefficient 0.378

Cohesion 0.10 MPa

Modulus of Elasticity 2375 MPa

Shear Modulus 47 MPa





1.2.2 Basic Mechanical Properties of Block

Masonry:

It also included tests on masonry prism and masonry

panels, three samples for each, for the estimation of

compression, shear and diagonal tension strength,

Young and shear elastic moduli. Table 1 reports the

basic mechanical properties of block masonry

investigated in the present research study.

1.2.3 Observed Response of Walls (Before and

after Retrofitting):

All of the walls have shown a flexure response at the

initial stages whereby cracks where developed at the

base of the wall which propagated with increasing

lateral displacement till rocking of the wall is initiated.

However, inclined shear cracks were also developed

before pure rocking of wall may be achieved. Sliding is

also observed at the bed joints of wall. The final

damaged wall included severe inclined cracks in

masonry units and cracks at the head and bed joints. The

retrofitted walls developed flexure cracks at the base

which is propagated to ensure rocking of the wall.

Significant inclined cracks were observed at the

compressed toe of wall, which could finally lead to toe

crushing and possibly simultaneous overturning of the

wall.

Performance-Based Damage Scale:

For all the walls, the retrofitting technique changed the

response mechanism from a flexure-shear mixed type

response to a flexure-toe crushing type response, when

reinforced plaster was applied on both faces of the wall.

The mechanism is not changed when reinforced plaster

was applied on one face of the wall only. Shoaib et al.

(2011) reported the lateral force-displacement response

of tested walls (before and after retrofitting), where it

was observed that the retrofitting technique can increase

the lateral strength by ten percent and increases the

ductility (the ratio of ultimate displacement capacity to

yield displacement) by 100 percent due to ensuring

flexure response of walls. Increase in the idealized yield

displacement capacity of the retrofitted wall relative to

the counterpart unretrofitted wall was not significant.

In the present study, the following damage scale is

developed (after FEMA, 2003) in view of the trend

towards deformation-based seismic design and

assessment of block masonry structures (before ad after

retrofitting).

where θi represents the mean drift limit states i.e. drift

threshold value; θy represents the idealized yield drift

limit derived from the bi-linearization of lateral force-

displacement response of wall; θu represents the ultimate

drift limits. The indicated damage state and performance

levels are attained upon the exceedance of the specified

drift limit. The approximated observed θy is 0.11% for

unretrofitted wall and 0.13% for retrofitted wall while θu

is 0.62% for unretrofitted wall and 1.25% for retrofitted

wall.

2 Example Structures for Investigation:

2.1 General Characteristics of Block Masonry

Structures in Pakistan:

The block masonry construction in Pakistan consists of

load-bearing walls of 150mm to 200mm thickness,

using single block unit (block size 300mm length,

200mm width, 150mm thickness) i.e. single-leaf,

constructed in cement-sand mortar mix ratio of 1:6. The

blocks are generally manufactured from combination of

cement, sand, and crushed stone with a mix proportion

(by volume) of 1:6:12 to 1:8:14. This mix proportion

results in 80 to 100 block units for 50 kilograms of

cement (one bag) with significantly low compressive

strength (Naseer et al., 2010). However, the current

recommendations for reconstruction suggest block units

of compressive strength more than 10 MPa, which

cannot be achieved looking at the current practice in the

field.

These structures are provided with 130mm to 150mm

thick reinforced concrete slab, as also common for brick

masonry construction type in urban exposure. Light

wooden and steel roof truss with Galvanized Iron sheet

and wooden floors are also practiced now in rural and

urban parts of the country. The clear inter-story height

and ground floor height ranges roughly between 2.0m to

3.0m and 2.5m to 3.5m, respectively. The current

construction guidelines recommend the use of band

beams at plinth/ lintel/ roof levels in order to reduce the

aspect ratio of wall, avoid the out-of-plane bending of

walls and ensuring in-plane integrity of structures.





2.2 Example Structures: Geometric Detailing and

Mathematical Modelling for Analysis:

Figure 2 shows typical plan of structures analyzed in the

present research study. Such construction practice can

be found in the major seismic areas of Pakistan e.g. the

northern parts of the country. However the current

construction recommendation i.e. symmetric detailing

for construction to avoid the torsional inertia in the

structures subjected to ground motions, is followed for

case study investigation.

In the present investigation, one of the structures (left) is

considered with floor area of 50 square meter and wall

density of 4.20 percent in the excitation direction, while

the second with floor area of 75 square meter and wall

density of 4.60 percent. Since for a given masonry

typology, these parameters significantly influence the

response of structures (Ahmad et al., 2011a) in which

the wall density has relatively high impact on the

seismic performance of structures.

Direction of Excitation Direction of Excitation

Plan View

Back Elevation

Front Elevation

Plan View

Back Elevation

Front Elevation

Figure 2: Typical Geometric Detailing of Example Structures Investigated in the Present Study. From Left to Right:

ubm1 (Left) and ubm2 (Right) in Case of Unretrofitted Structures which are Labeled as ubm1r (Left) and ubm2r

(Right) in Case of Retrofitted Structures

The structures are considered with interstorey height of

3.50m at ground floor and 3.00m at first floor and

provided with 1.50m deep spandrels. The height of

windows and doors opening is considered as 1.00m and

2.00m respectively. In the case study structures UBM1

& UBM1R, the length of corners short piers is 1.00m,

the middle piers have length of 1.50m while the

intermediate piers are 0.90m in length. For UBM2 &

UBM2R structures the length of corners short piers is

1.30m, the middle piers have length of 1.90m while the

intermediate piers are 1.15m in length. The structures

were first designed with the basic material properties of

block masonry. In the second phase, the material

properties of retrofitted walls were considered for the

design.

The equivalent frame modelling technique proposed by

Magenes and Fontana (1998) and Kappos et al (2002) is

used for the mathematical modelling and nonlinear

dynamic time history analysis (NLTHA) of case study

structures. This technique can be efficiently extended

for vulnerability assessment of structures using fully

dynamic approach. In the present study, the analysis is

carried out using OpenSees (McKenna et al., 2010).

In this approach, the masonry walls are idealized as one

dimensional beam-column elements with bending and

shear deformation with infinitely stiff joint element

offsets at the ends of the pier and spandrel elements.

The beam-column element is completely defined by

masonry Young modulus, shear modulus, wall sectional

area and wall moment of inertia. Each element is

assigned with a nonlinear lateral force-displacement





constitutive law depending on the response mechanism

of wall. Considering the ultimate damage mechanisms

of masonry wall, after Calvi and Magenes (1997) and

Tomazevic (1999); the diagonal shear, flexure (toe

crushing), and shear sliding analytical models are

employed for lateral strength evaluation of a given wall:

where Vf represents the lateral strength for flexure (toe

crushing) mechanism; D represents the length of wall; t

represents the thickness of wall; p=P/(Dt) represents the

mean vertical stress due to axial load P; HP represents

the height of wall; ψ is 1.0 for a cantilever pier and 0.5

for a pier with fixed-fixed boundary conditions; fu

represents the compressive strength of masonry; k is a

coefficient used to idealize the stress distribution at the

compressed toe of wall (Magenes et al., 2000),

considered as 0.85; Vd represent the shear strength for

diagonal shear damage mechanism; ftu is the diagonal

tensile strength; b=1 for HP/D≤1, b=HP/D for

1<HP/D<1.5 and b=1.5 for 1.5≤HP/D; VS represents the

sliding shear strength; µ and c represent the coefficient

of friction and cohesion of masonry as global strength

parameters.

A correction factor is proposed (Magenes and Calvi,

1997; Mann and Muller, 1982) to transform the local

parameters obtained using triplet/couplet test to global

parameters. The above shear strength models have been

discussed earlier in Magenes and Calvi (1997) and

Tomazevic (1999) for appropriate applications.

Due to the provision of reinforced concrete slab, deep

spandrels, ring beam above the wall and band beam at

lintel level (in order to ensure strong-spandrel and

weak-pier condition), the response of spandrel is

approximately considered as elastic, whereas inelastic

response is considered only in the piers.

Figure 3 shows the comparison of lateral shear strength

obtained experimentally and calculated using the above

analytical models. The models shows reasonable

performance in terms of providing estimate of lateral

strength and predicting response mechanism. In case of

retrofitted walls the increase in masonry compressive

strength (24%), diagonal tension and shear strength

(74%) due to retrofitting are considered accordingly in

the strength evaluation.

Unreinforced Wall Retrofitted Wall0

5

10

15

20

Tested Walls

Lat

eral

Str

eng

th (

To

n)

Shear Sliding Model

Diagonal Shear Model

Toe Crushing Model

Experimentally Observed

GoverningGoverning

Figure 3: Comparison of Lateral Shear Strength

obtained Experimentally and Calculated using

Analytical Shear Strength Models. From Left to Right:

Unreinforced Wall and Retrofitted with Reinforced

Plaster

In case of shear sliding model the friction coefficient

component is not increased due to the fact that

experimental data is not available to define µ for

reinforced plastered wall. Due to this the shear sliding

model will result in lower estimate of lateral strength for

retrofitted wall. This situation also hinders the reliable

application of sliding shear model (modified Mohr-

Coulomb shear model). However from the experimental

tests, it is observed for the considered retrofitting

technique that, although diagonal shear cracks may

occur in wall, toe crushing is always the governing

mechanism. Thus, only the diagonal shear and toe

crushing models are used confidently for retrofitted wall

in the present study.

The indicated increase in the mechanical properties of

masonry due to wire-mesh application is typical for the

considered reinforced plaster design scheme. Thus,

generalization of lateral strength model (considering the

modified mechanical properties) to retrofitted walls with

different geometry is reasonably possible. However, it is

worth to mention that the effect of reinforced plaster on

the mechanical characteristic depend significantly on

the strength and thickness of mortar used for plastering,

gauge spacing of wire mesh (the amount of

reinforcement), mesh orientation, the quality of bond at

the interface of reinforced plaster and wall and spacings

of connecting screws (ACI 549R-97, 1997; ElGawady

et al., 2004; Shah, 2011).

For the present case study investigation, the shear

strength model was governing in all cases for the

unreinforced structures while the strength obtained

using the toe crushing model is used for the retrofitted

case. The diagonal shear strength model always

provided higher estimate of lateral strength for

retrofitted walls. The nonlinear behavior, lateral force-

displacement response of frame element is idealized as

bi-linear with hysteretic response for masonry walls in





unretrofitted case, as recommended by Ahmad et al.

(2011b) for shear mechanism of masonry walls. Similar,

constitutive law is also used recently by Menon and

Magenes (2011) for seismic analysis of masonry

structures. The retrofitted walls showed flexure

response with toe damage for which bi-linear rocking

re-centering rule is approximately used. Figure 4 shows

typical force-displacement rule for shear and flexure

mechanism of masonry walls which are employed for

unretrofitted and retrofitted cases, respectively, in the

present study.

Figure 4: Force-Displacement Constitutive Law for In-

Plane Lateral Response of Walls, after Ahmad (2011b):

From Left to Right: Force-Displacement Idealization

for Shear Mechanism (Approximately Extended to

Unretrofitted Walls) and Force-Displacement

Idealization for Flexure Mechanism (Approximately

Extended to Retrofitted Walls)

3 Performance Evaluation of Block Masonry

Structures:

3.1 Nonlinear Dynamic Reliability Based Seismic

Assessment Method (NDRM):

The study included a nonlinear dynamic reliability

based approach for probabilistic-based seismic

performance assessment of case study structures. This

includes the incremental dynamic analysis (IDA)

technique for structural analysis as proposed by

Vamvatsikos and Cornell (2002), to obtain demand

(drift demand) on a structure in a probabilistic fashion

which is convolve with the probabilistic capacity (drift

capacity) to obtain the exceedance probability of

specified damage state (e.g. collapse). Figure 5

describes graphically the probabilistic-based assessment

framework.

Generally, the exceedance probability of specified

damage state for a given ground motion is obtained

using the classical reliability formulation that includes

the integration of joint probability density function of

demand and capacity, which is nevertheless

inconvenient to evaluate numerically. The present study

thus employed the First Order Reliability Method

FORM approximations (Der Kiureghian, 2005; Pinto et

al., 2004) to obtain the damage state exceedance

probability. This procedure has been recently

investigated against moderate and large magnitude

earthquakes for damage and collapse assessment of

existing structures in Pakistan (Ahmad, 2011) which

shows reasonable performance of the approach.

The case study structures are analyzed dynamically

using NLTHA with ten natural accelerograms extracted

from the PEER NGA data base for stiff soil site and

inter-mediate field condition i.e. sites within 15km to

30km of fault rupture (see Figure 6 and Table 2).

The selected accelerograms are compatible with the

Pakistan building code (BCP, 2007) specified

acceleration response spectrum for Type D soil and EC8

(CEN, 2004) specified response spectrum for Type C

soil of NEHRP classification. The accelerograms are

linearly scaled to multiple levels of shaking intensity in

order to derive response curve (drift demand correlated

with shaking intensity) for a given structure which are

employed to calculate the probability of exceedance of

specified limit state given the shaking intensity.

3.2 Collapse Assessment of Case Study Structures:

In the present study the collapse limit state corresponds

to the damage level when structure is forced to near-

collapse or complete collapse state. This damage level

occurs when collapse prevention (CP) is exceeded as

specified for the performance-based assessment and

design of structures. The physical damage corresponds

to the state of structure on the verge of partial and total

collapse where the occupants may be injured and

structural damages are not economically repairable.





Figure 5: Probabilistic Framework for Seismic Assessment of Structures, after Ahmad (2011). From Left to Right

and Top to Bottom: Mathematical Modelling of Structure, Selection and Scaling of Accelerograms, Derivation of

IDA Curves for Structures and Estimation of Exceedance Probability of Ground Motions for Specified Performance

Level

Figure 6: Typical Shape of Acceleration Response

Spectra with the Associated Uncertainties in Spectral

Acceleration and Comparison with Code Specified

Spectra

In the present study, the tested unretrofitted walls were

deformed into life safety performance range till the peak

load was developed, strength and stiffness degradation

is noticed and when repair to damage was feasible. This

means that the wall was pushed in to life safety

performance range but the CP limit state was not

passed. The ductility and drift limit computed in this

case is lower than the actual CP limit states but which is

nevertheless conservative for assessment purpose. The

retrofitted walls were also deformed into the life safety

performance range while CP was not reached, however

significant plasticity was observed in the wall. Also,

significant damage was observed at the compressed

corners which may lead to toe crushing and possibly

simultaneous overturning of the wall.

Table 2: Characteristics of Acceleration Time Histories

Selected for NLTHA, after Menon and Magenes (2011)

and Pampanin et al. (2002)

Thus, the ultimate drift capacity used in the present

study is a conservative (relatively lower than the actual)

estimate for CP limit state. Thereby the structures are

presumably considered with reduced ultimate drift

capacity, which is nevertheless conservative.

Figure 7 shows the response curve, which shows the

interstorey drift demand for specified level of shaking

intensity, derived through IDA procedure. For each of

the target PGA, the drift demand is convolve with drift

capacity using the FORM approximation. The structure





reliability against the collapse is calculated (shown in

Figure 8) which can provide help on identifying the

critical ground motions with different confidence level.

The reliability curves show that the case study structures

(UBM1, UBM2, UBM1R and UBM2R) can survive

ground shaking intensity of 0.28g, 0.35g, 0.40g and

0.50g respectively whereas ground motions with

shaking intensity great than 0.55g, 0.73g, 0.73g and

0.90g respectively will cause the collapse of structures

(with 95% confidence). The intermediate shaking

intensity will have different level of confidence for

survival and collapse of structures.

Figure 7: Response Curves Derived for Case Study

Structures using IDA. The Curves Show the Interstorey

Drift Demand on Structures for Specified Ground

Motion

Figure 8: Reliability Curves Derived for Case Study

Structures using NDRM. The Curves Show the Chances

of Survival and Collapse of Structure for Specified

Ground Motion

The investigation shows that considering the survival of

structures against collapse, the reinforced plastering

technique increases the seismic capacity by 43% for

both case study structures which provide significant

increase in the ground shaking intensity that structure

can survive.

This indicates that similar block masonry structures

retrofitted with reinforced plaster can be used

confidently in high hazard zones of Pakistan like Zone 3

and Zone 4 specified by building code BCP (2007), in

order to ensure the safety of occupants during strong

ground shaking. The curves also show that the seismic

capacity of block masonry structures can be

significantly increased with increasing wall density. For

example a ten percent increase in the wall density

increased the seismic capacity by 25%, as observed in

the capacity of UBM2 relative to UBM1 structure.

3.3 Design Life Performance of Block Masonry

Structures:

The above discussion mainly considered the collapse

capacity of structures against earthquake induced

ground motions, considering the performance of

structures for a single earthquake scenario having

magnitude of about Mw 7 (0.26 std.) within 15km to

30km of fault rupture. Also, the above discussion is

limited to the performance of structures to ensure the

safety of occupants during a specified earthquake

events. Considering the fact that a structure may be

subjected to various frequent and rare event earthquakes

in its design life, which may cause different level of

damage in the structures due which repair will be

required for the structure at regular interval. This

situation also make necessary to assess the performance

of a structure for the intended economic losses it can

incur over the design life.

Figure 9: Fragility Functions Derived for Case Study

Structures using NDRM Method and the FEMA

Damage Scale. From Left to Right and Top to Bottom:

UBM1, UBM2, UBM1R and UBM2R

For this purpose fragility functions are derived for case

study structures (Figure 9) using the procedure outlined





in Ahmad (2011). Vulnerability curves are derived

convolving the structure fragility functions with the

economic consequence factor (ECF), where ECF relates

physical damage with the monetary loss for the required

repair as a fraction of total replacement cost of structure.

The damage scale, as developed earlier, and ECF

proposed by FEMA (2003) is used in the present study.

This damage scale requires the CP and Yield limit state

drift capacities to develop the whole damage scale, see

Eq. (1) to (4). The FEMA ECF specify repair cost ratio

of 0.02 for slight damage, 0.10 for moderate damage,

0.50 for heavy damage and 1.0 for near-

collapse/collapse structures.

0 0.5 1 1.50

0.2

0.4

0.6

0.8

1

Peak Ground Acceleration(g)

MD

R

Mea

n D

amag

e R

atio

UBM2

UBM1R

UBM2R

UBM1

Figure 10: Vulnerability Curves Derived for Case Study

Structures using FEMA Economic Consequence Factor.

The Curves Show the Ratio of the Cost of Repair to

Replacement (i.e. MDR) of Structure for Specified

Ground Motion

For the case study applications, the structures are

investigated in the northern parts of Pakistan where such

construction system is practiced abundantly, as

mentioned earlier. The District of Mansehra is

considered for application in the present study, which

also presents area with high seismic hazard.

0.01 0.1 10.00001

0.0001

0.001

0.01

0.1

1

10

Peak Ground Acceleration(g)

Annual

Pro

bab

ilit

y o

f E

xce

edan

ce

PMD Seismicity Model

GSHAP Seismicity Model

Figure 11: Seismic Hazard Curves Derived for

Mansehra District through PSHA using the PMD and

GSHAP Seismicity Models

The scope of this case study is to assess the block

masonry performance subjected to different possible

earthquakes in the region. The probabilistic framework

as outlined by Ahmad (2011) is used to assess structures

for the derivation of average annual loss (AAL). The

AAL represents the annual amount needed to be

arranged every year in order to be ready for the required

repair following any earthquake event in the region over

the design life of structure.

For this purpose all possible seismic sources around the

region is considered, for which possible ground motions

are generated using empirical ground motion prediction

equations. The ground motions are assigned with the

associated annual probability in order to derive the

hazard curve using the standard PSHA framework (see

Figure 11). In the present study three ground motion

prediction equations (Abrahamson and Silva, 2008;

Boore and Atkinson, 2008; Campbell and Bozorgnia,

2008) and two seismicity models (GSHAP after

Giardini et al., 1999; and PMD, 2007) are used to derive

hazard curve. The vulnerability and hazard curves are

convolved to estimate the AAL. Figure 12 shows the

AAL calculated for each of the case study structures.

UBM1 UBM2 UBM1R UBM2R0

0.5

1

1.5

2

Structure

Av

erag

e A

nn

ual

Lo

ss

Per

cen

tag

e o

f R

epla

cem

ent

Co

st

0.38

0.67

0.97

1.60

Figure 12: Average Annual Economic Losses (AAL)

Calculated for Case Study Structures in Mansehra

District. The AAL Specify the Amount (as a Fraction of

Replacement Cost) Required on Annual Bases for the

Regular Repair and Maintenance of Structures due to

Earthquakes

It is observed that UBM1, UBM2, UBM1R and

UBM2R structure will acquire AAL of 1.60, 0.97, 0.67

and 0.38 percent of the replacement cost of structure

respectively, resulting in 63, 103, 150 and 263 years of

useful life of structures. This indicates that the

retrofitting technique will reduce the economic losses of

both case study structures by 60 percent on annual bases

and will increase the useful life of structure by about

150 percent. The useful life of structure here represents

the time span when the structure accumulated economic

losses reaches the replacement cost.





4 Closure:

4.1 Summary:

The aim of the study was to evaluate the performance of

block masonry structures before and after retrofitted

with reinforced plaster in order to assess the feasibility

of cost-effective and indigenous retrofitting technique

for improving the seismic performance of masonry

structures. Investigation is carried on block masonry

construction system against earthquake induced site

amplified ground motions. It included incremental

dynamic analysis of example structures for the

derivation of seismic response curves which are

analyzed through FORM approximation in order to

derive the structure reliability curves. Furthermore,

fragility functions are also derived which are employed

to derive the vulnerability curves. Additionally, PSHA

is carried out to develop seismic hazard curves for

Mansehra District (a high seismicity region). The

economic loss the structure can incur on annual bases is

calculated convolving the site hazard with the structure

vulnerability curves. This also gave, inversely, the

useful life of unretrofitted and retrofitted structures. The

investigation showed significant good performance of

the reinforced plastering technique in improving the

capacity of block masonry structures against earthquake

induced shaking.

4.2 Conclusions:

The following conclusions are derived based on the

seismic analysis of block masonry structures that can

ensure in-plane lateral response to ground shaking (right

through the length of walls). The out-of-plane failure of

walls in these structures are avoided practicing

restricted wall aspect ratio (height-to-thickness ratio)

and ensuring good wall-to-wall and wall-to-floor

connectivity. The case study structures are considered

with rigid reinforced concrete floors provided with ring

beam (monolithically connecting walls to the floor),

band beam (at the lintel level) and deep spandrels

(ensuring strong-spandrel to weak-pier condition). The

in-plane walls are considered with inelastic response

while the spandrels are considered with elastic response.

The study considered the experimentally obtained

mechanical properties of block masonry in Pakistan. In

the case study structures, the lateral strength of

unreinforced wall is provided by shear damage

mechanism (sliding shear model, modified Mohr-

Coulomb strength model) while the counterpart

retrofitted wall provided lateral strength with flexure

(toe crushing) mechanism. The lateral strength of

retrofitted wall increased by ten percent while the

ductility and drift capacity increased by about 100

percent. These findings are also applicable to block

masonry structure of the above characteristics and the

reinforced plaster design scheme as mentioned.

• For both unretrofitted and retrofitted walls, the

analytical lateral shear strength models are comparable

in predicting the response mechanism of walls

investigated though in-plane quasi-static cyclic test. The

models (Mohr-Coulomb shear model in case of

unretrofitted wall and toe crushing model in case of

retrofitted wall) provided reasonable estimate of lateral

shear strength.

• The investigation on case study structures showed

that considering the survival of structures against

collapse, the technique increases the seismic capacity by

43% which can provide significant increase in the

ground shaking intensity the structure can survive (with

95% confident).

• Locating and investigating the structures in high

seismicity area showed that the technique will reduce

the economic losses by about 60 percent on annual

bases and will increase the useful life of structure by

about 150 percent.

• The example unreinforced block masonry structures

retrofitted with reinforced plaster can ensure the safety

of occupants during strong ground motion, thus can be

confidently used in moderate to high hazard zones, as

mentioned.

• The better seismic performance of case study

retrofitted structures is primarily governed by the

increase in ductility capacity of walls while less

contribution (almost negligible) is provided by ten

percent increase in lateral strength.

• Additionally, the investigation also showed that a

ten percent increase in the wall density increases the

seismic capacity by 25 percent, reducing the economic

losses by 40 percent and increasing the structure useful

life by 63 percent. This indicates that the wall density of

masonry structures is also a crucial parameter in

improving the seismic performance.

4.3 Future Developments:

• There is a need to further investigate the reinforced

plaster retrofitting technique for masonry walls of

different geometry and loading condition. It can help in

developing generalized lateral shear strength models

which can be employed then confidently to assess the

performance of retrofitted structures analytically.

• The present study investigated masonry structures

with toe crushing mechanism of retrofitted walls and

ductility 100 percent greater than the counterpart

unretrofitted case. Further investigated is required to

understand the response mechanism and ductility

capacity for various cases of walls varying the

geometric and loading conditions to generalize the

effect of reinforced plaster for structures with any

geometric detailing.

• There is a need to further investigate retrofitted

walls with various geometries and reinforced plaster





design schemes to deduce deformation-related

performance objectives, defining different level of

required repair with the associated cost of repair, in

retrofitting techniques in view of the trend towards

deformation-based seismic design of structures.

• The present case study investigation considered bi-

linear hysteretic rule for shear damage mechanism

(unretrofitted case) and bi-linear centering rule for toe

crushing mechanism (retrofitted case). However,

investigation is required in this regard which may

further improve the findings reported herein and which

can in turn increase the confidence of approach in future

applications.

Acknowledgement:

The authors are grateful to Prof. Jason Ingham of the

University of Auckland, New Zealand; Prof. Joanna M.

Dulinska of the University of Cracow, Poland and Asst.

Prof. Paloma Pineda of the University of Seville, Spain

for their efforts in reading the manuscript in detail,

kindly encouraging the research study and providing us

with constructive comments and remarks which were

very helpful in improving the manuscript.

References:

[1] Abrahamson, N. and Silva, W. (2008), Summary of

the Abrahamson and Silva NGA ground--motion

relations, Earthquake Spectra, 24(1); 67-97.

[2] ACI 549R-97. (1997). State-of-the-art report on

ferrocement. ACI Committee Report/ACI 549R-97,

American Concrete Institute (ACI), Detroit, USA.

(http://www.bpesol.com/bachphuong/media/images

/book/549r_97.pdf)

[3] ADB-WB. (2005). Pakistan 2005 earthquake:

Preliminary damage and needs assessment.

Technical Document, Asian Development Bank

and World Bank, Islamabad, Pakistan.

(http://www.adb.org/Documents/Reports/pakistan-

damage-needs-assessment.pdf)

[4] Ahmad, N. (2011). Seismic risk assessment and

loss estimation of building stock of Pakistan. PhD

Thesis, ROSE School-IUSS Pavia, Pavia, Italy.

[5] Ahmad, N., Ali, Q., Ashraf, M., Naeem, K. and

Alam, B. (2011a). Seismic structural design codes

evolution in Pakistan and critical investigation of

masonry structures for seismic design

recommendations. International Journal of Civil,

Structural, Environmental and Infrastructure

Engineering Research and Development; 1(1): 42-

85. (http://www.epernicus.com/shared_documents

/1139/ijcerd-civil-engg-Seismic_Structural_

Design_ Codes_Evolution_-Naveed-Ahmad.pdf)

[6] Ahmad, N., Crowley, H., Pinho, R. and Ali, Q.

(2011b). Frame-elements constitutive law for

nonlinear static and dynamic analyses of masonry

buildings. In Cheung, S. O., Yazdani, F., Ghafoori,

N. and Singh A. (eds.): Modern Methods and

Advances in Structural Engineering and

Construction; Research Publishing Service,

Singapore. (http://rpsonline.com.sg/proceedings/

9789810879204/html/978-981-08-7920-4_S2-

S63.xml)

[7] Badrashi, Y., Ali, Q. and Ashraf, M. (2010).

Reinforced concrete buildings in Pakistan. EERI

Housing Report, World Housing Encyclopedia,

Oakland, CA, USA. (http://www.world-

housing.net/WHEReports/wh100184.pdf)

[8] BCP-2007. (2007). Building Code of Pakistan–

Seismic provision 2007. Technical Document,

Ministry of Housing and Works, Islamabad,

Pakistan. (http://www.pec.org.pk/downloadables/

buildingCode/Building Code of Pakistan.zip)

[9] Boore, D.M. and Atkinson, G.M. (2008), Ground–

motion prediction equations for the average

horizontal component of PGA, PGV, and 5%–

damped PSA at spectral periods between 0.01s and

10.0s. Earthquake Spectra, 24(1); 99-138.

[10] Calvi, G.M. and Magenes, G. (1994). Experimental

results on unreinforced masonry shear walls

damaged and repaired, Proceedings of Tenth

International Brick and Block Masonry Conference,

Calgary, Canada.

[11] Campbell, K. and Bozorgnia, Y. (2008), NGA

ground motion model for the geometric mean

horizontal component of PGA, PGV, PGD and 5%

damped linear elastic response spectra for periods

ranging from 0.01 to 10s. Earthquake Spectra;

24(1): 139-171.

[12] CEN, (2004). Eurocode 8: Design of structures for

earthquake resistance, Part1: General rules-seismic

actions and general requirements for buildings, EN

1998-1-3, CEN, Brussels, Belgium.

[13] Der Kiureghian, A. (2005). First– and second–order

reliability methods. In Nikolaidis, E., Ghiocel, D.

M. and Singhal, S. (eds.) Engineering design

reliability handbook.

[14] ElGawady, M., Lestuzzi, P. and Badoux, M.

(2004). A review of conventional seismic

retrofitting techniques for URM. Proceedings of the

Thirteenth International Brick and Block Masonry

Conference, Amsterdam. (http://imacwww.epfl.ch/

GenieParasismique/EDOC_ST09/Course_6/old/89

ElGA%2013-IBMaC%20state%20of%20the

%20art.pdf)

[15] ERRA. (2006). Earthquake reconstruction and

rehabilitation in earthquake affected areas.

Earthquake Reconstruction and Rehabilitation

Authority (ERRA), Islamabad, Pakistan.

[16] FEMA. (2003). Multi–hazard loss estimation

methodology, earthquake model. HAZUS-MH

Technical Manual, Federal Emergency





Management Agency (FEMA), Washington, DC,

USA.

[17] Giardini, D., Grunthal, G., Shedlock, K.M., and

Zhang, P. (1999), The GSHAP: global seismic

hazard map. Annali Di Geofisica, 42(6), 1225-

1230.

[18] Hamid, A., Mahmoud, A. and Abo El Maged, S.

(1994). Strengthening and repair of unreinforced

masonry structures: state-of-the-art. Proceedings of

Tenth International Brick and Block Masonry

Conference, Calgary, Canada.

[19] Magenes, G. and Calvi, G.M. (1997). In-plane

seismic response of brick masonry walls,

Earthquake Engineering and Structural Dynamics,

26 (11), 1091-1112.

[20] Magenes, G. and Fontana, D. (1998). Simplified

non-linear seismic analysis of masonry buildings.

Proceedings of the British Masonry Society, 5(8),

190-195.

[21] Magenes, G., Bolognini, D. and Braggio, C. (2000).

Metodi semplificati per l’analisi sismica non lineare

di edifici in muratura (in Italian), GNDT, Rome,

Italy.

[22] Mahmood, H. and Ingham, J.M. (2011). Seismic

vulnerability assessment of Pakistan unreinforced

masonry buildings at a national scale.

Seismological Research Letters; 82(5), 676-685.

[23] Mann, W. and Muller, H. (1982). Failure of shear-

stressed masonry-an enlarged theory, tests, and

application to shear walls. Proceedings of the

British Ceramic Society 27, 223-235.

[24] Maqsood, S.T. and Schwarz, J. (2008). Seismic

vulnerability of existing building stock in Pakistan.

Proceedings of the Fourteenth World Conference

on Earthquake Engineering. Beijing, China.

[25] McKenna, F., Fenves, G.L. and Scott, M.H. (2010).

Open system for earthquake engineering simulation

(OpenSees) version2.1.1, University of California,

Berkeley, CA, USA. (http://opensees.berkeley.edu)

[26] Menon, A. and Magenes, G. (2011). Definition of

seismic input for out-of-plane response of masonry

walls: I. parametric study. Journal of Earthquake

Engineering; 15(2), 165-194.

[27] Naseer, A., Naeem, A., Hussain, Z. and Ali, Q.

(2010). Observed seismic behavior of buildings in

northern Pakistan during the 2005 Kashmir

earthquake. Earthquake Spectra; 26(2): 425-449.

[28] Pampanin, S., Christopoulos, C., and Priestley, M.

J. N. (2002). Residual deformations in the

performance-based seismic assessment of frame

structures. Technical Report, IUSS Press, Pavia,

Italy.

[29] Pinto, P.E., Giannini, R., and Franchin, P. (2004).

Seismic reliability analysis of structures. IUSS

Press, Pavia, Italy.

[30] PMD. (2007), Seismic hazard analysis and zonation

for Pakistan, Azad Jammu and Kashmir. Technical

Report, Pakistan Meteorological Department

(PMD), Islamabad, Pakistan.

[31] Rossetto, T. and Peiris, N. (2009). Observations of

damage due to the Kashmir earthquake of October

8, 2005 and study of current seismic provisions for

buildings in Pakistan. Bulletin of Earthquake

Engineering; 7(3):681-699.

[32] Schuller, M., Atkinson, R. and Borgsmiller, J.

(1994). Injection grouting for repair and retrofit of

unreinforced masonry. Proceedings of Tenth

International Brick and Block Masonry Conference,

Calgary, Canada.

[33] Shahzada, K. (2006). Field practicing manual:

Guidelines for standard construction (English

Translation). Technical Document, Earthquake

Engineering Center, Peshawar, Pakistan.

(http://iisee.kenken.go.jp/net/seismic_design_code/

pakistan/field_practicing_manual.pdf)

[34] Shah, A.A. (2011). Applications of ferrocement in

strengthening of unreinforced masonry columns.

International Journal of Geology; 5(1), 21-27.

(http://www.naun.org/journals/geology/20-080.pdf)

[35] Shoaib, M., Naseer, A., Shahzada, K., Naeem, A.

and Ashraf, M. (2011). Earthquake disaster

mitigation through retrofitting of unreinforced

concrete block masonry buildings. Advanced

Material Research; 255-260,2627-2631.

[36] Stephenson, M. (2008). Notes from experience in

post-earthquake rural housing reconstruction in

Pakistan. Proceedings of the Workshop on Build

Back Better, Beijing, China. (http://www.un.org.cn/

public/resource/9330387be56a506bac9cae9aef6d54

00.pdf)

[37] Tomazevic, M. (1999). Earthquake-resistant design

of masonry buildings-innovation in structures and

construction Vol.1, Imperial College Press,

London, UK.

[38] Vamvatsikos, D. and Cornell, C. (2002).

Incremental dynamic analysis. Earthquake

Engineering and Structural Dynamics, 31(3), 491-

514.

[39] Kappos, A.J., Penelis, G.G. and Drakopoulos, C.G.

(2002). “Evaluation of simplified models for lateral

load analysis of unreinforced masonry buildings”,

Journal of Structural Engineering, 128(7), 890-897.

Seismic Performance Evaluation of Reinforced Plaster...

Documents

Transcript of Seismic Performance Evaluation of Reinforced Plaster...