Posada Evaluation of Seismic

7/29/2019 Posada Evaluation of Seismic

http://slidepdf.com/reader/full/posada-evaluation-of-seismic 1/13

1

EVALUATION OF THE SEISMIC PERFORMANCE OF PRECAST INDUSTRIAL

BUILDINGS IN TURKEY

Mauricio Posada* and Sharon L. Wood**

*Graduate Research Assistant, Ferguson Structural Engineering Laboratory, University of Texas,10100 Burnet Road, Building 177, Austin, TX 78758; [email protected]**Professor of Civil Engineering, Ferguson Structural Engineering Laboratory, University of Texas, 10100 Burnet Road, Building 177, Austin, TX 78758; [email protected]

Abstract

Precast frame buildings are used throughout Turkey for industrial facilities. One-story

warehouses are the most common structural configuration; however, low-rise commercial andmanufacturing facilities are also constructed using precast concrete members. These structuralsystems are economical to construct and provide large open areas needed for manufacturing.Many precast industrial building collapsed during the recent earthquakes in Turkey. This papersummarizes relationships between the observed damage in one-story warehouse structures andthe stiffness of the lateral-load resisting system.

Introduction

Precast construction was introduced in Turkey in the 1960s. During the 1990s, approximately90% of the warehouse and light industrial facilities were constructed using precast members

(Karaesmen, 2001). The most common structural system for these facilities is based on astructural configuration that was developed in Western Europe to carry gravity loads only (Ersoyet al. 1999). Turkish engineers modified the connection details so that the precast buildings havethe capacity to resist lateral loads. However, each producer of precast elements has developed aunique set of connection and reinforcement details, and the details vary appreciably fromproducer to producer.

Structural damage and collapse (Fig. 1) of precast buildings was widely reportedthroughout the epicentral regions of the 1999 earthquakes in Turkey (Ataköy, 1999; EERI,2000). The objective of this investigation was to document the observed damage and determinethe likelyÃ p h r Ã Ã S r r h p u r Ã s Ã u r Ã V v r v Ã s Ã U r h Ã F p h r y v Ã V v r v Ã 7 ÷ h v o v Ã

University, Middle East Technical University, Purdue University, and the University of

Minnesota visited more than 50 precast industrial buildings in the epicentral regions of theAugust 1999 Kocaeli and November 1999 Düzce earthquakes. Their observations, and theresults of a parametric study to identify the causes of the observed structural damage, aresummarized in this paper.

Topics for future collaborative research related to precast construction were discussed byresearchers, designers, and precast producers during a US-Turkey seminar in Ankara in June2001. The paper concludes with a summary of these research needs.



2

Characteristics of Precast Industrial Buildings

Wide variations in the characteristics of precast construction were observed. The buildings

shown in Fig. 2 are typical of one-story warehouse facilities that were constructed throughoutnorthwest and central Turkey. The multi-story buildings shown in Fig. 3, which included spacefor offices and manufacturing, tended to be concentrated in the industrial areas around ø h i y Ã Ã

Construction costs, reported in 2000 US dollars, ranged from $7/m2 for the single-storywarehouses to $35/m2 for the multi-story structures. The structural systems in both types of buildings shared a number of common features, as discussed below.

Because the one-story industrial buildings represent the most common form of precastconstruction in the epicentral region and the overwhelming majority of the structures thatsustained damage during the 1999 earthquakes, the majority of this paper is devoted to studyingthe seismic response of these structural systems.

One-Story StructuresThe single-story buildings tended to be rectangular in plan with one to four bays in the transversedirection and ten to thirty bays in the longitudinal direction. Transverse bay widths ranged from10 to 25 m, and longitudinal bay widths ranged from 6 to 8 m. Story heights also ranged from 6and 8 m.

A v t r Ã ) Ã Ã Q u t h u Ã s Ã 9 h h t r q Ã Q r p h Ã 7 v y q v t Ã r h Ã 6 q h h h Õ

Figure 2: Photographs of Typical Single-Story Warehouse Facilities



3

Figure 4 shows an elevation of a typical one-story building with two bays in thetransverse direction. The columns are supported by precast socket footings, and are assumed tobe fixed at grade level. Flexural hinges were observed at the base of columns in many buildings,

which confirms this assumption. No evidence of foundation rotation or damage was observed.Long-span roof girders are oriented along the transverse axis of the building. The depth

of these girders often varies along their length, forming the triangular shape shown in Fig. 4.Beams with U-shaped cross sections are oriented along the longitudinal axis of the building.These beams function as gutters to collect water from the roof. Purlins span between the roof girders at regular intervals. Typically five to eight purlins ran between adjacent roof girders.

In one-story construction, the precast girders, gutter beams, and purlins were pinned atboth ends. Vertical dowels extended up from the supporting member and the horizontalelements were cast with vertical holes near their ends to accommodate these dowels. The holeswere filled with grout in most buildings. In some cases the dowels were threaded, and nuts wereinstalled before grouting. Typically, lightweight materials, such as metal decking or asbestos

panels, were used to form the roof. Clay tile infill was used in most cases for the exterior walls,but precast concrete wall panels were also used.

The typical, one-story industrial building depends entirely on the cantilevered columnsfor lateral strength and stiffness. Even when precast wall panels were used for cladding, theconnection details were developed such that the wall panels did not contribute to the lateralstiffness of the building.

Figure 3: Photographs of Multi-Story Precast Facilities

Figure 4: Transverse Elevation of Typical Precast Warehouse

Long-Span GirdersGutter Beams

Columns



4

Three types of structural damage were frequently observed in the one-story industrial

buildings: flexural hinges at the base of the columns; axial movement of the roof girders whichled to pounding against the supporting columns or unseating of the roof girders; and out-of-planemovement of the roof girders which led to tilting of the beams and rotation off the supports.

Multi-Story Structures

The multi-story precast buildings also tended to be rectangular in plan. The roof was formedwith the same long-span girders, gutter beams, and purlins that were used for single-storystructures. All of these connections were pinned, as discussed previously. The columns wereprecast as a single member, so there were no splices along the height. Socket footings were usedto support the columns, and the base of each column was assumed to be fixed.

The floor beams were supported on the column corbels, and hollow-core planks were

used to form the intermediate floors. A thin topping slab was cast over the planks. Althoughprecast members were used to construct the intermediate floors, the connections between thecolumns and the beams (in both directions) were designed to be moment resisting. Theseconnection details varied appreciably among the precast producers. However, field welding wascommon. Often the intermediate floors were present in only portions of the building, and otherareas were open, as shown in Fig. 6.

Figure 5: Transverse Elevation of Multi-Story Precast Industrial Building

Figure 6: Photograph of Interior of Multi-Story Industrial Facility



5

Building Code Provisions

The current Turkish Building Code (1998) uses the structural behavior factor, R, to convertelastic spectral accelerations to design spectral accelerations. The code defines four structuralsystems for cast-in-place reinforced concrete buildings, four systems for precast concretebuildings, and four systems for structural steel buildings. The structural behavior factors rangefrom 8 for reinforced concrete or structural steel moment-resisting frames to 4 for precastconcrete shear walls. The structural system used for the typical one-story industrial buildingsdescribed in the previous section is assigned a structural behavior factor of 5. The design baseshear also depends on the effective peak ground acceleration, the intended use of the building,the soil characteristics at the site, and the period of the building.

Elastic response spectra corresponding to the four soil categories identified in thebuilding code are shown in Fig. 7. The effective peak ground acceleration used to calculate thesespectra corresponds to the zone of highest seismic risk in Turkey.

With the exception of buildings with very short periods, the design spectral accelerationsare determined by dividing the elastic spectral accelerations shown in Fig. 7 by the structuralbehavior factor. Inter-story drift ratios are calculated using the lateral forces corresponding tothe design spectral accelerations and must not exceed the limits given below:

( )0035.0max ≤

∆

i

i

h(1)

( ) Rhi

i 02.0max ≤∆ (2)

where (∆i)max is the maximum inter-story displacement, hi is the height of the correspondingstory, and R is structural behavior factor. For typical one-story industrial buildings, themaximum calculated story drift ratio is controlled by Eq. 1 and is limited to 0.35%. Aninvestigation of four damaged one- Ã i v y q v t Ã v Ã 6 q h h h Õ Ã 6 h x | Ã ( ( ( Ã v q v p h r q Ã u h Ã h y y Ã

four buildings failed to satisfy this stiffness criterion.

0.0 1.0 2.0 3.00.0

0.2

0.4

0.6

0.8

1.0

1.2

Period, sec

S p e c t r a l A c c e l e r a t i o n , g

Z3

Z4

Z2

Z1

0.5 1.5 2.5

Z3

Z4

Z2

Z1

0.5 1.5 2.5

Figure 7: Elastic Response Spectra for Seismic Zone 1

(Damping Factor = 0.05)



6

Influence of Column Stiffness on Structural Performance

A parametric study was conducted to relate the behavior of one-story precast industrial buildings Ã u r Ã p y Ã v s s r Ã Ã 6 Ã ' Ã i Ã ! Ã Ã i v y q v t Ã v Ã 6 q h h h Õ Ã u v p u Ã h v r q Ã y v t u Ã q h h t r Ã

during the Kocaeli earthquake, was selected as the prototype structure for this study. The

transverse bay widths were 20 m, the longitudinal bay widths were 7.5 m, and the story heightwas 7 m.A linear model of the framing system in the transverse direction was developed. The

base of each column was fixed and the connections between the columns and roof girders werepinned such that vertical loads and shear were resisted, but the flexural resistance at the ends of the beams was negligible.

For the purpose of the parametric study, column dimensions were varied from 40 by40 cm to 80 by 80 cm. These dimensions correspond to the smallest and largest precast columnsthat were observed in the epicentral region. The cross-sectional dimensions and mass of thegirders, gutter beams, purlins, roofing materials, and cladding in the prototype building wereused in all analyses. The variation of the calculated fundamental period with the assumed

column dimensions is given in Table 1.

Table 1: Column Sizes Considered in Parametric Study

Column DimensionsDepth Width

CalculatedPeriod

cm cm sec40 40 1.1040 45 1.0545 40 0.9345 45 0.8850 40 0.80

50 45 0.7650 50 0.7350 55 0.7055 50 0.6455 55 0.6160 55 0.5460 60 0.5265 65 0.4570 70 0.4080 80 0.31

Ground Motion Records

Each of the buildings considered in the parametric study was analyzed using fifteen groundmotion records (Table 2). Most of the recording stations were within 50 km of the epicenters of the 1999 earthquakes, and all were within 20 km of the surface trace of the faults (EERI, 2000).The ground motion records were divided into two groups depending on the soil conditions at therecording station. Records from Bolu, Düzce, and Yarimca were used to determine the spectral



7

p u h h p r v v p Ã s Ã s Ã v y Ã u v y r Ã r p q Ã s Ã 6 p r y v x Ã B r i r Ã ø v Ã h q Ã T h x h h Ã r r used to

determine the spectral characteristics for stiff soil sites.

Elastic acceleration and displacement response spectra corresponding to a damping ratioof 2% were calculated for each ground motion record. Mean, maximum, and minimum spectraare plotted in Fig. 8 and 9 for the soft soil and stiff soil/rock sites, respectively. Although

statistical information is not shown in the plots, the maximum and minimum values weretypically less than 1.2 standard deviations from the mean value for the range of periods

considered.

Table 2: Ground Motions Considered in the Parametric Study

Peak

Acceleration

Epicentral

DistanceStation Component

g km

Soil Conditions

180 0.41Düzce (DZC)

270 0.5110** Soft Soil

090 0.23ø v Ã D a U

180 0.1712* Rock

240 0.30Yarimca (YPT)

330 0.3222* Soft Soil

Sakarya (SKR) 090 0.41 35* Stiff Soil

000 0.74Bolu (BOL)

090 0.8142** Soft Soil

000 0.27

Gebze (GBZ) 270 0.14 50* Stiff Soil

000 0.21Arcelik (ARC)

090 0.1360* Stiff Soil

180 0.32Düzce (DZC)

270 0.37110* Soft Soil

*Approximate distance to epicenter of Kocaeli earthquake.

**Approximate distance to epicenter of Düzce earthquake.

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

0.5

1.0

1.5

2.0

2.5

3.0

Period, sec

S p e c t r a l

A c c e l e r a t i o n , g Soft Soil Sites

Max

Mean

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

0.5

1.0

1.5

2.0

2.5

3.0

Period, sec

S p e c t r a l


Max

Mean

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

0.5

1.0

1.5

2.0

2.5

3.0

Period, sec

S p e c t r a l


Max

Mean

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00

10

20

30

40

50

60

Period, sec

S p e c t r a l D i s p l a c e m e n t , c m Soft Soil Sites

Mean

Max

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00

10

20

30

40

50

60

Period, sec


Mean

Max

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00

10

20

30

40

50

60

Period, sec


Mean

Max

Min

Figure 8: Mean, Maximum, and Minimum Values of Elastic Response Spectra for

Ground Motions Recorded on Soft Soils (Damping Factor = 0.02)



8

In general, the minimum spectral displacements from the soft soil sites were

approximately equal to the maximum spectral displacements from the stiff soil/rock sites. For

periods less than 0.25 sec, the maximum spectral accelerations from the stiff soil/rock sites were

nearly the same as the maximum spectral accelerations from the soft soil sites. However, for

periods greater than 0.5 sec, the maximum spectral accelerations from the stiff soil/rock siteswere approximately equal to the minimum spectral accelerations from the soft soil sites.

Displacement Capacity of Idealized Buildings

The displacement capacity of the idealized columns was determined by first calculating the

moment-curvature response of each of the cross sections listed in Table 1. Reinforcement ratiosof 1, 2, and 3% were used in the analysis. The assumed arrangement of the longitudinal

reinforcement is shown in Fig. 10.

The behavior of the concrete was modeled using the stress-strain relationship developed

by Hognestad (1951), a concrete compressive strength of 30 MPa, an initial modulus of elasticityof 26,000 MPa, and a limiting compressive strain of 0.0035. The stress-strain relationship forsteel was assumed to be linear to the yield point. The yield plateau was assumed to extend to a

strain of 0.01, and strain hardening was considered for strains between 0.01 and 0.10. The yieldstress of the steel was assumed to be 420 MPa, the tensile strength was 500 MPa, and the

modulus of elasticity was 204,000 MPa. The contribution of the transverse reinforcement inconfining the column core was ignored because ties with 90-degree hooks were used throughout

the epicentral region.

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

0.5

1.0

1.5

2.0

2.5

3.0

Period, sec

S p e c t r a

l A c c e l e r a t i o n , g Stiff Soil/Rock Sites

Max

Min

Mean

0.0 0.5 1.0 1.5 2.0 2.5 3.00.0

0.5

1.0

1.5

2.0

2.5

3.0

Period, sec

S p e c t r a

l A c c e l e r a t i o n , g Stiff Soil/Rock Sites

Max

Min

Mean

0.0 0.5 1.0 1.5 2.0 2.5 3.00

10

20

30

40

50

60

Period, sec

S p e c t r a l D i s p l a c e m e n t , c m Stiff Soil/Rock Sites

Mean

Max

Min

0.0 0.5 1.0 1.5 2.0 2.5 3.00

10

20

30

40

50

60

Period, sec

S p e c t r a l D i s p l a c e m e n t , c m Stiff Soil/Rock Sites

Mean

Max

Min

Figure 9: Mean, Maximum, and Minimum Values of Elastic Response Spectra for

Ground Motions Recorded on Stiff Soils (Damping Factor = 0.02)

40% As

40% As

20% As

40% As

10% As

40% As

5% As

5% As

r = 1% r = 2% and 3%

Figure 10: Assumed Arrangement of Reinforcing Bars for Parametric Study



9

Because the cantilevered columns provided all the lateral stiffness for the one-storybuildings, the yield displacement and displacement capacity could be calculated using Eq. 3

and 4.

3

2 L y

y

⋅=∆φ

(3)

( )

−⋅⋅−+∆=∆

2

p

p yu yu L"

"

φ φ (4)

where y

∆ is the yield displacement at the roof, y

φ is the yield curvature,u

∆ is the displacement

capacity at the roof,u

φ is the curvature capacity, L is the height of the one-story building, and

p"

is the height of the equivalent plastic hinge. The plastic hinge length was assumed to be one-

half the depth of the cross section for all cases (Moehle, 1992).

Comparison of Drift Demand and Capacity

The inelastic displacement demands in the one-story buildings during the 1999 earthquakes were

approximated using the elastic displacement demands calculated using a damping factor of 2%(Shimazaki and Sozen, 1984). The elastic displacement demands from the fifteen ground motion

records are compared with the calculated yield displacements in Fig. 11. Similarly to Fig. 4 and5, mean, maximum, and minimum response spectra are plotted for the two groups of ground

motions. All displacements are plotted in terms of the drift ratio: the roof displacement dividedby the building height.

The data in Fig. 11 indicate that all of the buildings considered in the parametric study

had the ability to resist the displacements induced by the composite maximum response spectrumfor the stiff soil/rock sites without yielding. However, the data in Fig. 11 also indicate that

idealized buildings with calculated periods larger than 0.7 sec would yield when subjected to themean response spectrum for the soft soil sites and all the idealized buildings would yield when

subjected to the composite maximum response spectrum for the soft soil sites. These resultswere not sensitive to the amount of longitudinal reinforcement in the columns.

Figure 11: Comparison of Drift Demand with Calculated Drift at Yield

0.00.0 0.2 0.4 0.6 0.8 1.0 1.2

1.0

2.0

3.0

4.0

5.0

Period, sec

D r i f t R a t i o ,

%

Calculated Drift at Yield

r = 1.0%

r = 2.0%

r = 3.0%

Mean

Mean

Stiff Soil

Soft Soil



10

The calculated drift capacities of the idealized buildings are compared with the responsespectra for soft soil sites in Fig. 12. Idealized buildings with periods greater than 0.8 sec are

likely to be pushed beyond their displacement capacity by the mean ground motion at the softsoil sites. Only buildings with periods less than 0.5 sec are likely to survive the maximum

composite response spectrum without reaching their displacement capacity.

As indicated in Table 1, a period of 0.8 sec corresponds to a building with 50 by 40-cmcolumns. The overwhelming majority of the buildings visited in the epicentral region were

constructed with columns this size or smaller. A period of 0.5 sec corresponds to a building with60 by 60-cm columns. Less than 5% of the single-story buildings visited had columns larger

than this size.

The calculations summarized in this section agree with observations from the field. The

i v y q v t Ã v r v r Ã v Ã B r i r Ã h q Ã 6 q h h h Õ Ã r r Ã r Ã v v y h Ã v Ã r Ã s Ã p p v Ã h y v Ã Ã

The buildings in Gebze appeared to be undamaged, while a large number of buildings collapsedv Ã 6 q h h h Õ Ã Ã U u r Ã r h u h x r Ã q r h q Ã h Ã y v x r y Ã Ã i r Ã v t v s v p h y Ã q v s s r r Ã v Ã u r Ã Ã p v v r Ã

due to the soil conditions. Buildings that experienced satisfactory performance when founded onstiff soil were likely to collapse when founded on soft soil deposits.

Comparison with Turkish Building Code

Each of the idealized buildings was also analyzed using design response spectra defined in the

Turkish Building Code (1998) for soil classifications Z3 and Z4 (Fig. 13). A period of 0.7 sec

corresponded to the stiffness at which the idealized building located on a site with soft soilconditions (Z4) would satisfy the drift criterion in the building code (Eq. 1). The critical periodis increased to approximately 0.78 sec if the idealized building is located on a site with Z3 soil

conditions. The data shown in Fig. 12 indicate that buildings with fundamental periods in thisrange would likely experience drift levels near capacity for the mean response spectrum, and

would be pushed beyond capacity for the composite maximum spectrum for soft soil. Thiscomparison indicates that complying with the drift provisions in the Turkish Building Code is

not sufficient to control the damage observed during the 1999 earthquakes.

Figure 12: Comparison of Drift Demand from Soft Soil Sites

with Calculated Drift Capacity

0.0 0.2 0.4 0.6 0.8 1.0 1.20.0

1.0

2.0

3.0

4.0

5.0

Period, sec


%

Calculated Drift Capacity

r = 1.0%

r = 2.0%

r = 3.0%

0.0 0.2 0.4 0.6 0.8 1.0 1.20.0

1.0

2.0

3.0

4.0

5.0

Period, sec


%



11

Conclusions

This paper has focused on one aspect of the seismic behavior of precast, industrial buildings inTurkey: the flexural response of the transverse frames in one-story warehouses. Large

variations in the performance of precast industrial buildings were observed in the epicentralregions of the 1999 Kocaeli and Düzce earthquakes. Column dimensions and connection details

are considered to have a critical influence on the performance of this type of structure. Becausethe structural system is not redundant, inelastic action in any column can lead to unseating of the

roof girders and collapse of the roof. Drift must be controlled for this type of structural systemto reliably survive future earthquakes. The design provisions in the Turkish building code and

typical column dimensions observed in the epicentral region do not appear to be sufficient tocontrol damage on soft soil sites.

Future Collaborative Research

In June 2001, researchers from the University of Texas, University of Minnesota, Middle East

Technical Unive v Ã F p h r y v Ã V v r v Ã h q Ã 7 ÷ h v o v Ã V v r v Ã r Ã v u Ã q r v t r Ã h q Ã

producers from the Turkish Precast Concrete Association to discuss research results and future

needs. Four topics were identified for possible study: development of methods for rehabilitatingthe large inventory of vulnerable precast buildings, development of hybrid moment-resisting

connections for multi-story precast frames, development of structural systems that rely onprecast concrete panels to resist lateral forces, and development of design provisions for

reinforced masonry in Turkey.Research related to rehabilitation was not viewed as a high research priority, because the

practitioners felt that it would be extremely difficult to convince the owners of the buildings that

rehabilitation was needed. Unless a factory owner has direct experience of sustaining earthquakedamage, they will not want to invest further in an operational facility. If the factory is located

within the epicentral region of the 1999 earthquakes and was not damaged, then the owner isunlikely to believe the building is vulnerable to damage during a future earthquake.

0.0 0.2 0.4 0.6 0.8 1.0 1.20.00

0.25

0.50

0.75

1.00

Period, sec


%Z4

Z3

Drift Limit for Design

Figure 13: Elastic Drift Demands Calculated using Design Response Spectra



12

Development of moment-resisting frame connections was given a high priority. Anumber of producers have worked with researchers at METU in the past decade to improve their

proprietary connection details. However, most involve field welding, which is difficult toinspect, and is therefore a quality control concern. Some of the hybrid connections that we

developed in the US during the PRESSS research program were considered to be potentially

applicable in Turkey, with connections that rely on unbonded post-tensioned reinforcement beingparticularly attractive.

Development of ductile precast wall systems was also given a high priority. Precast wallpanels are typically not mixed with precast frame members, but the use of walls would add

stiffness to the structural systems. Innovative details developed during the PRESSS researchprogram may also provide a means of dissipating energy within the system.

One practical method for improving the seismic performance of one-story precast

structures is to construct the interior partitions and exterior walls using reinforced masonry,rather than clay brick infill. This change would have little influence on the operations within the

building, but could increase the lateral strength and stiffness dramatically. Reinforced masonryis currently not permitted by the Turkish Building Code; therefore, additional research is needed

to develop the appropriate design provisions.Each of these ideas represents a potential topic for future collaborative research. At least

one construction company has expressed an interest in participating in these activities, and iswilling to donate materials and construct large-scale test specimens.

Acknowledgments

U u r Ã h v h p r Ã s Ã V ÷ Ã @ Ã @ u h Ã F h h r r Ã B r Ã g p r i r Ã h q Ã U ÷ y Ã U h x Ã H v q q y r Ã

@ h Ã U r p u v p h y Ã V v r v Ã ù r x r Ã g q r Ã F p h r y v Ã V v r v Ã T h v Ã 6 q Ã F Õ y Õ o Ã 7 ÷ h v o v Ã

University), Julio A. Ramirez and Mete A. Sözen (Purdue University), Catherine W. French andArturo E. Schultz (University of Minnesota), Michael E. Kreger and Eric B. Williamson

(University of Texas), the Turkish Precast Concrete Association, and numerous engineers atB g F Ã Q r x v h ú Ã T r Ã 7 r h Ã h q Ã ` h Õ Ã H r x r v Ã h r Ã t h r s y y Ã h p x y

edged.

The work described in this paper sponsored by the US National Science Foundation

under grant CMS-0085337. The opinions expressed in this paper are not necessarily those of thesponsor.

References

Atak öy, H. (1999). “17 August Marmara Earthquake and the Precast Concrete Structures Builtby TPCA Members,” Turkish Precast Concrete Association, Ankara, Turkey.

Code for Buildings to be Built in Disaster Areas (1998). Ministry of Construction, Ankara,

Turkey. (In Turkish).

Earthquake Engineering Research Institute (2000). “1999 Kocaeli, Turkey, EarthquakeReconnaissance Report,” Earthquake Spectra, Supplement A to Vol. 16.

Ersoy, U., T. Tankut, and G. Özcebe (1999). “Damages Observed in the Precast Framed

Structures in the 1998 Ceyhan Earthquake and their Rehabilitation,” Department of CivilEngineering, Middle East Technical University, Ankara, Turkey.



13

Hognestad, E. (1951). “A Study of Combined Bending and Axial Load in Reinforced ConcreteMembers,” Bulletin 399, University of Illinois Engineering Experiment Station, Urbana,Illinois.

Karaesmen, E. (2001). “Prefabrication in Turkey: Facts and Figures,” Department of CivilEngineering, Middle East Technical University, Ankara, Turkey.

Moehle, J.P. (1992). “Displacement-Based Design of RC Structures Subjected to Earthquakes,” Earthquake Spectra, 8(3):403-428.

Shimazaki, K. and M.A. Sözen (1984). “Seismic Drift of Reinforced Concrete Structures,” Research Reports, Hazama-Gumi Ltd., Tokyo. (In Japanese).

Posada Evaluation of Seismic

Documents

Transcript of Posada Evaluation of Seismic