Performance-based seismic design of an irregular tall building – a case study

Ali Ruzi Ozuygur¨

PII: S2352-0124(15)00107-1
DOI: doi: 10.1016/j.istruc.2015.10.001
Reference: ISTRUC 62

To appear in:
Received date: 25 July 2014 Please cite this article as: Ozuygur¨
Revised date: 2 September 2015 Ali Ruzi, Performance-based seismic design of an irregular tall
Accepted date: 2 October 2015 building – a case study, (2015), doi:
10.1016/j.istruc.2015.10.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a
service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT

Performance-based seismic design of an irregular tall building – a case

study
Ali Ruzi Özuygur*1
Department of Structural Engineering, YPU – Yapı Proje Uygulama Co., Istanbul, Turkey

SUMMARY

Structural design of a 50-story tall reinforced concrete residential building, which was planned to be
constructed in Istanbul and given up afterwards by the investor, has been completed in accordance with the
draft version of Seismic Design Code for Tall Buildings in Istanbul that adopts performance-based seismic
design as the basic approach as Tall Buildings Initiative Guidelines do. Seismic design of the building has
formed the main part of the structural design process due to high seismicity of the proposed location and
extremely irregular floor plan not conforming to usual tall building structures. The building consists of two
individual buildings linked through stronger link slabs at top 13 stories whereas relatively weak slabs at
lower stories. The building has been designed for design basis earthquake by elastic response spectrum
analysis and its seismic performance has been checked for maximum considered earthquake by nonlinear
time history analyses carried out using PERFORM-3D.

KEY WORDS: performance-based seismic design, nonlinear time history analyses, tall building, irregular floor plan,
PERFORM-3D

1. INTRODUCTION
Application of performance-based seismic design (PBSD) to tall buildings is relatively new although its
history goes back to 1980s. With the issue of PBSD recommendations such as Recommendations for the
Seismic Design of High-rise Buildings by Council on Tall Buildings and Urban Habitat (CTBUH, 2008),
An Alternative Procedure for Seismic Analysis and Design of Tall Buildings Located in the Los Angeles
Region by Los Angeles Tall Buildings Structural Design Council (LATBSDC, 2008), Requirements and
Guidelines for the Seismic Design and Review of New Tall Buildings using Non-Prescriptive
SeismicDesign Procedures by Structural Engineers Association of Northern California (SEAONC, 2007),
and finally Guidelines for Performance-Based Seismic Design of Tall Buildings by Tall Buildings Initiative
Guidelines Working Group (TBI, 2010), many buildings especially located in high seismicity regions have
been designed using PBSD method around the world (Klemencic et al., 2006; Willford and Smith, 2008;
Çelebi et al., 2013). As expressed in these documents, PBSD has many advantages over traditional
prescriptive code-based design methods whose regulations are not fully suitable for tall buildings for their
unique structural behavior. Traditional design codes, briefly, a) are basically prepared to regulate the design
of low and medium rise buildings whose first translational mode is taken into account in seismic analysis;
b) have application limitations with regard to building height; c) impose rigid rules on the analysis and
structural system; d) prescribe elastic analysis with seismic force reduction factor which was widely
investigated by various researchers (Nassar and Krawinkler, 1991; Paulay and Priestley, 1992; Miranda
and Bertero, 1994; Whittaker et al., 1999; Elnashai and Mwafy, 2002; Thuat, 2014) in order to account for
the inelastic behavior of the buildings under major earthquakes, although it cannot be theoretically defined
especially for tall buildings due to their unique structural behavior. Unlike these drawbacks of prescriptive
code-based design, PBSD makes it possible to more realistically obtain displacements and accelerations of
stories, effects of higher modes and redistribution of shear forces of tall buildings in inelastic behavior
range (Klemencic et al., 2007; Moehle et al., 2011).

Few codes in the world have regulatory requirements towards PBSD of tall buildings. Seismic Design Code
for Tall Buildings in Istanbul (SDCTBI, 2008) was proposed in 2008; however it hasn’t been put into
implementation yet. SDCTBI adopts PBSD method and has provisions like those of previously mentioned
documents. Design objectives in SDCTBI are briefly stated as a) negligible damage and immediate
occupancy performance level under earthquake with 50% probability of exceedance in 50 years (return
period of 72 years) entitled D1 and service level earthquake (SLE); b) controllable damage and life safety
performance level under earthquake with 10% probability of exceedance in 50 years

* Correspondence to: Ali Ruzi Özuygur, YPU – Yapı Proje Uygulama, Morbasan Sok. No.10, Balmumcu, Beşiktaş, Istanbul,
Turkey. E-mail: aruzi@ypu.com.tr, aliruzi@gmail.com

ACCEPTED MANUSCRIPT

(return period of 475 years) entitled D2 and design basis earthquake (DBE); c) extensive damage and
collapse prevention performance level under earthquake with 2% probability of exceedance in 50 years
(return period of 2475 years) entitled D3 and maximum considered earthquake (MCE).

Various studies were conducted to investigate and advance the application of PBSD to tall buildings
especially in the last 10 years. Case studies (Moehle et al., 2011; Jones, 2011; Tuna, 2012) conducted by
Pacific Earthquake Engineering Research Center, University of California, Berkeley (PEER) aiming at
defining the seismic performance of tall buildings designed by alternative means have been among the most
important works in this regard so far.

In this paper, structural design procedure of the proposed building whose first version was presented by
Özuygur (2015) completed in accordance with the regulations of SDCTBI along with the recommendations
of other guidelines is presented. The building has been designed for DBE by elastic response spectrum
analysis and its performance has been checked for MCE by nonlinear time history analyses; and brief
analyses results and observations have been summarized.

2. STRUCTURAL DESIGN OF THE BUILDING BY LINEAR ANALYSIS

2.1 The Structural System

The residential building whose architectural render is given in Figure 1 and which is planned to be
constructed in Bomonti district of Istanbul has 50 stories above and two additional stories below grade.
Total height of the building from foundation level is 198 m with 3.8-m story height above grade and 4-m
story height below grade. The building has extremely irregular structural floor plan as shown in Figure 2
imposed by rigid architectural requirements which is not usually suitable for tall building structures. The
structural floor plan is antimetric about the axis 13. The vertical load bearing system of the building consists
of concrete slabs sitting on beams supported by shear walls and columns. The lateral load carrying system
of the building consists of shear walls with coupling beams distributed in floor plan as required by
architectural needs. Dimensions of the structural elements of lower stories are summarized in Table 1.
General slab thickness is selected as 0.16 m by iterative analysis of vibration and long term deflection under
sustained loads. The slab of corridor area between shear walls is selected as 0.3 m in order to increase
lateral stiffness of the building. The structure can be considered as two individual buildings (Building A
and Building B) linked through weak corridor slabs (link slab) at most of the stories (Figure 2a) and fully
continuous floor slab (link slab) at top 13 stories (Figure 2b). The slab thickness of link area and adjacent
spans at top 13 stories is selected as 0.3 m considering in-plane forces of the slab caused by different
dynamic behavior of the buildings under seismic forces. Thicknesses of the shear walls are 0.8 m at lower
stories and gradually reduced to 0.4 m at top stories. Thickness of the shear walls at the axis 11 and 15 is
0.8 m all over the height in order to support 18-m long-span beam at link area. Diameter of the column is
1.5 m at lower stories and gradually reduced to 0.8 m at top stories.
 ACCEPTED MANUSCRIPT

Figure 1. Architectural render of the building by Tago Architects.

ACCEPTED MANUSCRIPT

(a) Typical floor plan below level 183.30 m.

(b) Typical floor plan above level 187.10 m.

Figure 2.
Typical floor plan.
 ACCEPTED MANUSCRIPT

2.2 Materials and Vertical Loads

Selected concrete class is C50 with characteristic strength fc = 50 MPa and reinforcing steel is S420 with
yield strength fsy = 420 MPa. Superimposed dead load is taken as 3.5 kPa which counts for floor coating
and separating walls in all residential areas. Live load is taken as 2 kPa in all residential areas, 5 kPa in all
public and car parking areas. Line load of 3.5 kN/m is taken as cladding weight on the beams at exterior
line of each floor. Additional area load of 15 kPa is taken in mechanical areas based on the information
provided by the mechanical designer.

2.3 Site Specific Assessment of Seismic Hazard

A site specific Probabilistic Seismic Hazard Assessment Report (PSHAR) was prepared by Erdik et al.
(2011) to develop the DBE and MCE response spectra and spectrally matched time history pairs. Istanbul
has suffered from numbers of destructive earthquakes throughout history (Ambraseys and Finkel, 1991).
Seismic hazard of Istanbul and therefore of this building is caused by seismically active Marmara Fault
System which is the north branch of well-known North Anatolian Fault. One of fault segmentation models
used for the assessment of seismic hazard of Marmara region is given in Figure 3 (Erdik et al., 2004). All
these fault segments have generated destructive earthquakes throughout history with moment magnitudes
mostly larger than 7.0.

Figure 3. A fault segmentation model for Marmara region (Erdik et al., 2004).

A possible largest earthquake having Mw = 7.25 can be expected by rupture of 4 segments of Marmara Fault
System. Depth of hypocenter can be assumed to be 12 km based on previous earthquakes experienced at
North Anatolian Fault. The distance between the building location and Marmara Fault System is about 23
km. Having S wave velocity of 760 m/s (ELC, 2009), the soil under the foundation belongs to B/C class in
SDCTBI and National Earthquake Hazards Reduction Program of the U.S.A (NEHRP). PSHAR using
these earthquake characteristics gives peak ground accelerations (PGA) and spectral accelerations (Sa) at
short period (SS) and period of 1 second (S1) as shown in Table 2.
Uniform hazard response spectrum of each earthquake level for 5% damping, Fa = 1 and Fv = 1
(NEHRP/IBC/SDCTBI) as well as the elastic spectra for spectrally matched ground motion time histories
of MCE are given in Figure 4.
 ACCEPTED MANUSCRIPT

(a) (b) (c)

Figure 6. First three mode shapes of the building: (a) first mode; (b) second mode; (c) third mode.
Figure 7. Elastic soil pressure beneath the foundation.
Different structural response modification factors are used in accordance with the provisions of SDCTB I
2.5 Wind Design
and based on structural design objectives. R = 6 is used for almost all flexure calculations be cause of
expected ductile behavior; R = 2 is used for all shear calculations because of expected brittle behavior;
Wind tunnel tests have been conducted by RWDI (RWDI , 2013) in order to define wind loads, wind
and R = 1 is used for flexure, shear and axial (in-plane) force calculations of link slab and beams between
induced structural response, cladding pressures and pedestrian level wind microclimate conditions. Win d
Building A and Building B at top 13 stories in order to keep them elastic under the largest earthquake.
tunnel model is given in Figure 8. Wind-induced bas e shear is about 0.004 W, therefore seismic forces
dominate the structural design. Total lateral displacement due to the wind load of 50-year return period is
The internal forces obtained with R = 6 are scaled by a factor calculated as the inver se ratio of base shear
obtained as 0.045 m. As for the human comfort, the acceleration due to wind load at top story of the
to the minimum base shear given in SDCTBI as below:
building is obtained as 7.9, 11 and 12 milli-g for winds of 1, 5 and 10-year of return period respecti vely.
As shown in Figure 9, these acceleration values almost satisfy residential comfort requirements of RWDI
Vb,min = 0.04 SMS(DBE) W (1)
and ISO 10137.
where SMS = Fa SS is spectral acceleration for short period and W is the seismic weight of the building
calculated as the sum of all dead loads and 30% of live loads as specified in Turkish Earthquake Code
2007 (TEC, 2007).

Seismic weight of the building is calculated as W = 2169303 kN. The minimum base shear is calculated
as Vb,min = 0.0356 W.

Lateral seismic loads are taken into account in four different directions based on the irregular shape of the
building; and the base shear scale factor calculate d from the minimum base shear is about 1.5 for all
seismic load directions.

Maximum elastic soil pressure beneath the foundatio n is obtained about 1100 kPa as given in Figure 7;
which is smaller than the allowed elastic soil pres sure 1200 kPa given by the geotechnical engineering
consultant.

Figure 8. Wind tunnel model.

 ACCEPTED MANUSCRIPT

Figure 9. Peak acceleration values due to wind load.

2.6 Design Challenges

Structural analysis and detailing of outer beams of link slabs at top 13 floors shown in Figure 10 hav e
been one of the most important design challenges in this project. Besides the response spectrum analys is
by R = 1, additional torsional moments MA and MB shown in Figure 10 with all possible sign
combinations are applied in order to depict the wor st case the beams can face considering the assumed
different behavior of Building A and B. MA and MB are calculated as follows:

MA = MB = 0.5 x (Vi+1 - Vi) x 0.05 x L, (2)

th
where Vi is the story shear of i floor and L is the approximate length of Building A or B. The axial forces
have also been checked by the results of nonlinear time history analysis by MCE.

Figure 10. Torsional moments applied to the irregular floor.

3. PERFORMANCE-BASED DESIGN BY NONLINEAR TIME HISTORY ANALYSES

Nonlinear time history analyses have been conducted by DBE ground motions for the evaluation of life safety
and MCE ground motions for the evaluation of collapse prevention performance level utilizing
 ACCEPTED MANUSCRIPT

PERFORM-3D (CSI, 2011b). Only the results of nonlinear time history analyses by MCE ground motions and
performance evaluation for collapse prevention level are presented here.

3.1 Key Points of Nonlinear Modeling and Performance Evaluation

Nonlinear modeling features in PERFORM-3D are mostly referenced to (Powell, 2007; Ghodsi and Ruiz,
2010). Key points of nonlinear modeling features and performance acceptance criteria are summarized
below:

a) Expected strength is used for concrete and reinforcing steel as fce = 1.3 fc and fse = 1.17 fsy respectively.
b) 2% of damping is used in the analyses.
c) The effective stiffness values of structural members used for linear elastic analysis by DBE are used
herein as well.
d) All frame elements are modeled as FEMA Beam and FEMA Column.
e) Outer beams of link slabs at top 13 floors are modeled as FEMA Column to high axial forces.
f) Link slabs are modeled as elastic shell elements; and two separate diaphragms are assigned to Building
A and Building B so as to more realistically obtain in-plane forces of link slab and axial forces of
beams at link slab area.
g) Boundary elements and the web of shear walls are modeled separately using their reinforcement ratios.
A considerable amount of modeling time has been saved by this way whereas the more amounts have
been consumed for running the model due to its size.
h) Being the basic references for SDCTBI as well as other guidelines, FEMA 356 and ASCE 41 are used
for the reference of performance acceptance criteria.

3.2 Results of Nonlinear Time History Analyses

Nonlinear time history analyses results such as lateral displacements and story drift ratios; compression and
tension strains, shear forces as well as axial forces of sample shear walls; chord rotation of sample coupling
beams are presented in this section to represent the overall seismic performance of the building under MCE.

Lateral displacement of lower left point at the floor levels and the corresponding story drift ratio are given
in Figure 11. Story drift ratio of this point is one of the maximum values and much smaller than the allowed
value 0.03 (LATBSDC, 2008; TBI, 2010; SDCTBI, 2008).

Maximum compression and tension strain of sample shear walls are given in Figure 12; and the peak
compression strain is smaller than the allowed value 0.018 (SDCTBI, 2008) for unconfined concrete noting
that the boundaries of the shear walls are properly confined in accordance with the relevant codes (TEC,
2007; ACI, 2011) which results in higher capacity of compression strain. Similarly, the peak tension strain
is smaller than the allowed value 0.06 (SDCTBI, 2008).

Shear force diagrams of sample shear walls are given in Figure 13. Shear forces obtained from nonlinear
time history analyses are compared with those obtained from linear elastic analysis by DBE with R = 2. It
is observed that the shear forces of linear elastic analysis are much smaller than that of nonlinear time
history analyses. Although the maximum shear stresses are under the allowed value of 0.83 (ACI,
2011) in both cases, extra amount of shear reinforcement are needed for the case of nonlinear time history
analyses. It is also observed that the shear profiles of walls are generally not affected by coupling effect of
stronger link slabs at top 13 floors; in other words, no extinct jump is observed at top 13 floor levels. This
shows that the weaker link slabs at lower stories behave as diaphragm elements affecting (smoothing) shear
profile of walls; furthermore the Building A and B don’t vibrate as individual buildings under seismic
loadings.
Axial force diagrams of sample shear walls are given in Figure 14. Axial forces obtained from nonlinear
time history analyses are compared with that obtained from linear elastic analysis by DBE with R = 2. It is
observed that the axial forces of shear walls at outer boundary of irregular floor plan obtained from
nonlinear time history analyses are larger than that obtained from linear elastic analysis; and they exceed
the accepted axial force limit of 0.3 fce Acw (TBI, 2010) at lower stories. This result is probably caused by
Figure 14. Axial force diagram of sample shear walls.
The author sincerely appreciates the valuable contributions of Macit Yurtsever, Cemile Aydın, Ali Đhsan
Güner along with other work team (of YPU); Şamil Şeref Polat (consultant to SĐNPAŞ GYO, the investor),
Bora Çelik, Mesut Çağlarım (of SĐNPAŞ GYO); Đrfan Balıoğlu, Coşkun Kuzu (of Balkar

ACCEPTED MANUSCRIPT

Mühendislik, the peer review company); and Barış Erkuş and other distinguished consultants (of Istanbul
Technical University and Boğaziçi University, Turkey). The author exclusively thanks Şamil Şeref Polat
for his sincere guidance and help. The author also sincerely acknowledges the support of YPU as well as
SĐNPAŞ GYO and thanks Övünç Tezer for proofreading of the manuscript.

REFERENCES

ACI. 2011. Building Code Requirements for Structural Concrete and Commentary (ACI 318M-11). American
Concrete Institute: Farmington Hills, MI, 430.
Ambraseys NN, Finkel CF. 1991. Long-term seismicity of Istanbul and of the Marmara Sea region. Terra Nova.
3: 527-539.
Çelebi M, Huang M, Shakal A, Hooper J, Klemencic R. 2013. Ambient response of a unique performance-
based design tall building with dynamic response modification features. Struct. Design Tall Spec.
Build. 22, 816–829, DOI: 10.1002/tal.1093.
CSI. 2011a. ETABS, Extended 3D Analysis of Building Systems Software, Nonlinear Version 9.7.1. Computers
and Structures, Inc.: Berkeley, CA.
CSI. 2011b. PERFORM 3D, Nonlinear Analysis and Performance Assessment for 3D Structures User Guide,
Version 5. Computers and Structures, Inc.: Berkeley, CA.
CSI. 2012. SAP2000, Structural Analysis Program, Nonlinear Version 15.2.1. Computers and Structures, Inc.:
Berkeley, CA.
CTBUH. 2008. Recommendations for the Seismic Design of High-rise Buildings. Council on Tall Buildings and
Urban Habitat. Illinois Institute of Technology, Chicago, IL.
ELC. 2009. Soil investigation report for Sinpas Bomonti Residence. ELC Group Co. Istanbul, Turkey.
Elnashai AS, Mwafy AM. 2002. Overstrength and force reduction factors of multistorey reinforced concrete
buildings. Struct. Design Tall Spec. Build. 11, 329–351, DOI: 10.1002/tal.204.
Erdik M, Demircioğlu M, Şeşetyan K, Durukal E, Siyahi B. 2004. Assessment of Probabilistic Earthquake
Hazard in the Marmara Region. Soil Dynamics and Earthquake Engineering. Vol. 24, 605-631.
Erdik M, Demircioğlu M, Şeşetyan K, Yenidoğan C. 2011. Seismic hazard assessment for Sinpas Bomonti
Residence. Department of Earthquake Engineering, Bogazici University, Istanbul, Turkey.
Ghodsi T, Ruiz JAF. 2010. Pacific Earthquake Engineering Research/Seismic Safety Commission Tall Building
Design Case Study 2. Struct. Design Tall Spec. Build. 19, 197–256, DOI: 10.1002/tal.542.
Jones PS. 2011. Assessment of Performance Based Design Procedures for Tall Buildings. Ph.D. Dissertation.
University of California, Irvine.
Klemencic R, Fry AJ, Hooper JD, and Morgen BG. 2007. Performance-Based Design of Ductile Concrete
Core Wall Buildings-Issues to Consider Before Detailed Analysis. Struct. Design Tall Spec. Build.
16, 599–614, DOI: 10.1002/tal.437.
Klemencic R, Fry JA, Hooper JD. 2006. Performance-based design of tall reinforced concrete ductile core wall
systems. Struct. Design Tall Spec. Build. 15, 571–579, DOI: 10.1002/tal.383.
LATBSDC. 2008. An Alternative Procedure for Seismic Analysis and Design of Tall Buildings Located in
the Los Angeles Region. Los Angeles Tall Buildings Structural Design Council. Los Angeles, CA.
Miranda E, Bertero VV. 1994. Evaluation of strength reduction factors for earthquake-resistant design. Earthquake
Spectra 10(2): 357–379.
Moehle J, Bozorgnia Y, Jayaram N, Jones P, Rahnama M, Shome N, Tuna Z, Wallace J, Yang T, Zareian
F. 2011. Case Studies of the Seismic Performance of Tall Buildings Designed by Alternative Means.
PEER Report 2011/05. Pacific Earthquake Engineering Research Center, University of California,
Berkeley, CA.
Naish D, Wallace J, Fry JA, Klemencic R. 2009. Reinforced Concrete Link Beams: Alternative Details for
Improved Constructability. UCLA - SGEL Report 2009/06. University of California, Los Angeles.
Nassar AA, Krawinkler H. 1991. Seismic demands for SDOF and MFOD systems. Report No. 95, The John A.
Blume Earthquake Engineering Center, Stanford University, California.
Özuygur AR. 2015. Performance-based seismic design of an irregular tall building in Istanbul. Struct. Design Tall
Spec. Build. 24, 703–723, DOI: 10.1002/tal.1207.
Paulay T, Priestley MJN. 1992. Seismic Design of Reinforced Concrete and Masonry Buildings. John Wiley &
Sons: A Wiley Interscience Publications: Canada.
 ACCEPTED MANUSCRIPT

Powell GH. 2007. Detailed Example of a Tall Shear Wall Building. Computers and Structures, Inc.:
Berkeley, CA.
RWDI. 2013. Reports of Wind Tunnel Testing of Sinpaş Bomonti Tower. RWDI, U.K.
SDCTBI. 2008. Seismic Design Code of Tall Buildings in Istanbul. Draft Version. Department of
Earthquake Engineering, Bogazici University, Istanbul, Turkey.
SEAONC. 2007. Requirements and Guidelines for the Seismic Design and Review of New Tall Buildings
using Non-Prescriptive Seismic-Design Procedures. S tructural Engineers Association of Northern
California. San Francisco, CA.
TBI. 2010. Guidelines for Performance-Based Seismic Design of Tall Buildings. PEER Report 2010/05.
The TBI Guidelines Working Group, Pacific Earthquak e Engineering Research Center, University
of California, Berkeley, California.
TEC. 2007. Specifications for Structures to be Built in Disaster Areas (Turkish Earthquake Code).
Turkish Ministry of Environment and Urban Planning, Ankara.
Thuat DV. 2014. Strength reduction factor demands for building structures under different seismic levels.
Struct. Design Tall Spec. Build. 23, 42–53. DOI: 10.1002/tal.1018.
Tuna Z. 2012. Seismic Performance, Modeling and Fai lure Assessment of Reinforced Concrete Shear
Wall Buildings. Ph.D. Dissertation. University of California, Los Angeles.
Whittaker A, Hart G, Rojahn C. 1999. Seismic response modification factors. Journal of Structural
Engineering, ASCE, 125(4): 438–444.
Willford MR, Smith RJ. 2008. Performance Based Seis mic and Wind Engineering for 60 Story Twin
th
Towers in Manila. The 14 World Conference on Earthquake Engineering. Beijing, China.

AUTHOR'S BIOGRAPHY

Ali Ruzi Özuygur obtained his Ph.D. in structural e ngineering from Istanbul Technical University in 20 11. He works
for YPU – Yapı Proje Uygulama, a multi-disciplinary firm in Turkey, as structural engineering departme nt manager.
He has been involved in or managed structural design of numerous tall buildings and specialty structures in Turkey.