Journal of Earthquake Engineering

ISSN: 1363-2469 (Print) 1559-808X (Online) Journal homepage: https://www.tandfonline.com/loi/ueqe20

A Framework to Incorporate Probabilistic

Performance in Force-based Seismic Design of RC
Buildings as per Indian Standards

Prakash S. Badal & Ravi Sinha

To cite this article: Prakash S. Badal & Ravi Sinha (2020): A Framework to Incorporate
Probabilistic Performance in Force-based Seismic Design of RC Buildings as per Indian Standards,
Journal of Earthquake Engineering, DOI: 10.1080/13632469.2020.1713931

To link to this article: https://doi.org/10.1080/13632469.2020.1713931

Published online: 07 Feb 2020.

Submit your article to this journal

Article views: 41

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ueqe20
JOURNAL OF EARTHQUAKE ENGINEERING
https://doi.org/10.1080/13632469.2020.1713931

A Framework to Incorporate Probabilistic Performance in

Force-based Seismic Design of RC Buildings as per Indian
Standards
Prakash S. Badal and Ravi Sinha
Department of Civil Engineering, Indian Institute of Technology Bombay, Mumbai, India

ABSTRACT ARTICLE HISTORY

A framework for incorporating probabilistic seismic performance dur- Received 3 July 2019
ing prescriptive design of reinforced concrete buildings using Indian Accepted 5 January 2020
building codes is presented. The second-generation performance- KEYWORDS
based design procedure is considered. Thirty special RC frame build- Seismic design; response
ings have been numerically analyzed through nonlinear static and reduction factor;
time history analyses. Effects of seismicity, building height, bay-width, overstrength factor;
and number of bays on building performance are investigated. A performance-based design;
generalized expression for the overstrength factor is proposed. Indian design standards;
Expression for response reduction factor to achieve the desired probabilistic analysis
level of performance objective is also proposed. Low- and mid-rise
buildings in the high seismic region designed using current codes are
found to be deficient.

1. Introduction
A large number of Reinforced Concrete (RC) moment frame buildings that are designed
and constructed in India during the last few decades are based on prescriptive seismic
design and detailing standards of 1993 and 2002 (IS 13920 1993; IS 1893 (part 1) 2002). It
is estimated that between the decade of 2001 and 2011 alone, more than 4 million RC
tenements, that were required to comply with these standards, were constructed in India
(Census 2001, 2011). The Indian seismic standards are based on the force-based design
method, and the expected performance of buildings constructed using these standards are
not explicitly stated. As India has not experienced a major urban earthquake since the
widespread adoption of these standards, observational information on the performance of
these buildings is not available. The present paper quantifies the seismic performance of
code-compliant buildings under second-generation probabilistic performance-based
design framework. The paper also proposes seismic design parameters that can be used
with the force-based Indian seismic standards so that the envisaged seismic performance
objectives are achieved.
The seismic design philosophy for RC buildings permits limited damage during the
application of design-level force and relies on the ductile detailing of structural members
to develop energy dissipation capacity. Linear analysis is typically carried out wherein the
seismic design force is reduced by a response reduction factor (RRF). This factor

CONTACT Ravi Sinha [email protected] Department of Civil Engineering, Indian Institute of Technology
Bombay, Mumbai 400076, India
© 2020 Taylor & Francis Group, LLC
2 P. S. BADAL AND R. SINHA

represents the effects of a complex combination of nonlinear capacity of the structural

members, their buffer strength, and reduction in seismic demand due to nonlinear
behavior of building after damage in such a way that linear seismic analysis results can
be combined with the results from analysis for other loads to determine the design forces.
Based on the building response to monotonically increasing lateral load, Fig. 1 schema-
tically illustrates three seismic design parameters—response reduction factor, R, system
overstrength factor (OSF), Ω0, and deflection amplification factor, Cd—used for design
based on linear seismic analysis of a building (FEMA 450 2004). In this figure, VE is the
base shear corresponding to design earthquake assuming elastic behavior, Vdes is the design
base shear, Vmax is the base shear capacity of the building, ΔE is the roof drift corresponding
to VE assuming sufficient strength and stiffness of the building that allows it to remain
elastic, and Δdes is the design level of roof drift. Eurocode (EC 8 2004) utilizes similar
performance factors with different terminologies. The Indian seismic design standard,
considers Ω0 and Cd factors indirectly (IS 1893 (part 1) 2002). Table 1 presents a summary
of seismic design parameters for reinforced concrete moment-resisting frame (RCMRF)
buildings as recommended by IS 1893 (part 1) (2002), EC 8 (2004), and ASCE 7 (2010).
Initial studies to assess seismic design parameters focused on estimating the reduction in
seismic strength demand due to the nonlinear behavior of buildings (Krawinkler and Nassar
1992; Miranda and Bertero 1994; Whittaker, Hart, and Rojahn 1999). Mondal, Ghosh, and
Reddy (2013) assessed RRF value for four special RC frames of different heights (2-, 4-, 7-, and
12-story frames) designed as per Indian standards for seismic zone-IV (second highest seismic
hazard) using a deterministic framework and nonlinear-static analysis. For performance limit
corresponding to life safety, they reported inadequacy in the design of these frames in terms of
lower RRF values than the code-recommended value of 5.0. A relatively small number of
frames were considered in their study, which are not deemed as a representative set of
archetypical buildings as per FEMA P695 (2009). These studies were unable to consider the
probabilistic effects of variation of strength and deformation parameters and uncertainty in

Figure 1. Seismic design parameters used in structural design, based on FEMA 450 (2004).
 JOURNAL OF EARTHQUAKE ENGINEERING 3

Table 1. Seismic design parameters for RCMRF per IS 1893 (part 1) (2002), ASCE 7 (2010), and EC 8
(2004).
System Code R Ω0 Cd
Fully-ductile RCMRF IS1893 5.0 2.5a 5.0b
ASCE7 8.0 3.0 5.5
EC8c 5.9d,f - 5.9
Limited-ductile RCMRF IS1893 3.0 2.5a 5.0b
ASCE7 3.0 3.0 2.5
EC8c 3.9e,f - 3.9
a
Defined indirectly for soft-story shear amplification (similar to section 12.3.3.2 of ASCE 7–10).
b
Determined from the reduced level of MIDR of 0.4% (treating 2.0% as a benchmark from ASCE 7–10).
c
EC8 uses the terminology of behavior factor and displacement behavior factor for R and Cd, respectively.
d
For high ductility class (DCH).
e
For medium ductility class (DCM).
f
Factor αu/α1 taken as 1.30 for multi-story multi-bay structure.

the earthquake forces (Ellingwood 2001) and hence are not widely used any more due to
increased focus on performance-based seismic design.
In the current paper, a framework that translates the performance-based probabilistic
seismic design of buildings to the prescriptive design procedures while meeting code-envi-
saged seismic performance objective of Indian seismic standards is presented. Special RC
frame buildings have been considered for the lateral load-resisting system in the study. A set of
thirty archetypical building configurations designed and detailed per Indian standards (IS
13920 1993; IS 456 2000; IS 1893 (part 1) 2002) representing commonly found RC buildings in
urban India have been considered. These configurations have been decided so as to capture a
reasonably wide range of key design variables and structural attributes recommended in
published literature (FEMA P695 2009; Kircher et al. 2010). A probabilistic evaluation has
been carried out to ascertain their seismic behavior considering the effects of dynamic loading
and various types of uncertainties. Incremental dynamic analysis has been employed for
probabilistic evaluation of their seismic behavior. This methodology has been widely validated
for different lateral load-resisting systems (Han, Kim, and Baek 2018; Kircher et al. 2010;
Miyamoto et al. 2011). Seismic performance objective considered for the present study
corresponds to 10% probability of exceeding collapse prevention performance. Similar per-
formance objectives have been adopted in several studies (Franchin, Petrini, and Mollaioli
2018; Luco et al. 2007; Ulrich, Negulescu, and Douglas 2014).
The present study shows that the performance of some building configurations in seismic
zone-V that are designed as per Indian standards are not conservative and do not meet the
desired seismic performance, especially, for the low- and mid-rise buildings. The paper further
proposes a new expression of RRF value for special RC frame buildings so that the building
designs can meet the expected performance objectives. The study also shows that the OSF
reduces with severity in the seismicity and increase in the height of the building. A generalized
expression for OSF, as a function of seismicity and time period, is also proposed.

2. Probabilistic Seismic Performance Assessment

2.1. Seismic Performance Objectives
For the design of new buildings as well as evaluation of the existing ones, first-generation
performance-based design guidelines specify a matrix of expected performance levels of
4 P. S. BADAL AND R. SINHA

buildings corresponding to different levels of seismic demand (FEMA 273 1997; FEMA
356 2000; FEMA 445 2006). A part of this matrix is shown in Table 2. For existing
buildings, based on the rehabilitation objective, one or more of these performance
objectives can be targeted.
With the introduction of second-generation performance-based guidelines, seismic
performance objectives were modified from deterministic (e.g., a fixed value of demand
to capacity ratio) to probabilistic (e.g., specification of a probability of exceeding certain
seismic performance conditional on a hazard level) (FEMA 58-1 2012; FEMA P695 2009).
Consequently, the simple matching of a performance limit with a given hazard level, as
shown in Table 2, cannot suffice. Many early assessment studies for ordinary buildings are
based on an imprecisely-specified low probability of exceedance of collapse prevention
damage state for maximum considered earthquake (MCE) ground motion (Haselton and
Deierlein 2006; Ibarra and Krawinkler 2005). Luco et al. (2007) quantified the low
probability of exceedance as 10% probability level.
Targeting collapse prevention (CP) at MCE, i.e., cell p of Table 2, has been convention-
ally treated as a primary choice of performance objective for high seismic regions based on
the need to ensure the safety of the occupants (Haselton et al. 2008; ASCE 7 2010).
However, for moderate seismic regions in India, complete collapse of code-conforming
buildings due to earthquakes is relatively rare and therefore, not generally expected. To
reinforce this point, the foreword of seismic design code (IS 1893 (part 1) 2002) envisages
cell p to be the intended seismic performance. In view of the above discussion, seismic
performance objective of 10% probability of exceedance of collapse prevention at MCE
hazard level has been adopted in the present paper. It shall be noted that the seismic
performance objective in this paper has been chosen with a focus on the expectations of
the scientific and policy-making community specific to the region. Finally, performance
limits are based on commonly adopted system-level peak deformation parameter, max-
imum interstory drift ratio, MIDR (FEMA 356 2000; ASCE 41 2006).

2.2. Evaluation of Seismic Response Coefficients

From static pushover analysis, the overstrength factor (Ω0 ) of a building is defined as the
ratio of maximum base shear coefficient, Cs,max, to the design-level seismic response
coefficient, Cs (FEMA P695 2009). Mathematically,
Cs;max
Ω0 ¼ : (1)
Cs
Parabolic distribution of lateral load distribution along building height has been adopted
for pushover analysis as recommended in IS 1893 (part 1) (2002).

Table 2. A subset of performance objective matrix used in the current study (FEMA 356 2000; ASCE 41
2006).
Performance limit →
Hazard level ↓ IO LS CP
20%/50y f
10%/50y (DBE) j k
2%/50y (MCE) n o p
(k + p) constitutes basic safety objectives.
(k + p + any of f, j, or n) constitutes enhanced objectives.
 JOURNAL OF EARTHQUAKE ENGINEERING 5

Evaluation of response reduction factor (RRF) based on nonlinear static analyses is

unable to capture the influence of dynamic characteristics of the load and buildings, such
as dominant frequencies of the ground motions and natural frequencies of the building.
Additionally, constant redundancy factors suggested by the use of a statics-based approach
are unable to consider the probabilistic effects of variation of strength and deformation
parameters and uncertainty in the earthquake forces (Ellingwood 2001). FEMA P695
(2009) overcomes these shortcomings by proposing a risk-based approach for determining
seismic design parameters that correspond to a defined level of building performance
against seismic forces. Uncertainties in demand as well as in capacity are duly addressed in
this approach. This methodology has been used to evaluate the seismic design parameters
of various lateral load resisting systems in published literature (Denavit et al. 2016; Han,
Kim, and Baek 2018; Kircher et al. 2010; Miyamoto et al. 2011; Nobahar, Farahi, and
Mofid 2016; Sato and Uang 2010).
The procedure described in FEMA P695 (2009) uses incremental dynamic analysis
(IDA). The IDA is carried out by subjecting the structures to a series of scaled time
history excitations (Vamvatsikos and Cornell 2002). The FEMA P695 approach is based
on the concept of risk-consistency and is rooted in the PEER framework (Der Kiureghian
2005; Porter 2003). Figure 2 schematically shows the assessment of RRF using this
approach. For the building under consideration, fragility function for CP damage state
has been plotted between the value of intensity measure and the probability of exceedance
of CP.
In case of two-dimensional building models, the concept of controlling horizontal
component of the ground motion is used to consider the response along both directions
(Appendix F of FEMA P695). The controlling horizontal component is defined as the
component that induces the structure to reach a specific damage state at a lower value of
intensity measure. The geometric mean of spectral acceleration along two horizontal
direction, SageoM ðT Þ, has been used in this study as intensity measure to keep a constant
scaling factor for both components of the time history records (Whittaker et al. 2011).
Additionally, ground motion prediction models typically provide mean and standard
deviation for ln SageoM (Baker and Cornell 2011) and thus, the use of SageoM during
structural analysis brings homogeneity in the process.
Consider a building designed for seismic response coefficient, Cs. In the present article,
without loss of generalization, Cs is computed as per IS 1893 (part 1) (2002). Let SMT be
MCE spectral acceleration at its fundamental time period. For ordinary buildings, the ratio
of MCE spectral acceleration to the design-level seismic response coefficient is equal to
twice the RRF:
I SMT 2R
Cs ¼ SMT ) ¼ : (2)
2R Cs I
A factor of 2 in the numerator accounts for the definition of design-level ground motion,
which is defined as half of the MCE ground motion (IS 1893 (part 1) 2002). If any other
design standard-specific expression for Cs is used, above expression (2) shall be modified
accordingly. This, in turn, would modify the factor of 2R/I shown in Fig. 2.
Further, assuming a lognormal distribution of fragility, let μSa;CP and βTOT be its
parameters corresponding to collapse prevention obtained from the nonlinear time history
analysis of the building of interest. In order to meet the performance objective of “less
6 P. S. BADAL AND R. SINHA

Figure 2. Schematic representation of RRF for probabilistic seismic performance, based on FEMA P695
(2009).

than or equal to 10% probability of exceeding CP at MCE ground motion,” the following
inequality has to be satisfied:


ln SMT  ln μSa;CP  
Φ  10% ) SMT  μSa;CP exp βTOT Φ1 ð0:10Þ : (3)
βTOT
Substituting for SMT from (2) to (3), we get a lower bound on the RRF available to the
structure. Let us denote this lower bound as Ravl to distinguish it from RRF specified by
the design codes. Thus, we have
I μSa;CP  
Ravl ¼ exp βTOT Φ1 ð0:10Þ : (4)
2 Cs
The present study evaluates (4) to assess the intrinsic seismic capacity of a building in
order to meet the erstwhile defined seismic performance objective.

3. Archetypical Special RC Frame Buildings

3.1. Selection of Archetypical Buildings
In order to obtain a baseline behavior of special RC frame buildings, the current study
focuses on regular bare RC frames. A large number of distinct building configurations can
be found for a specific lateral load-resisting system. A set of index archetypical buildings
that adequately captures the key design variables is required for assessing the seismic
performance of a specific building typology (FEMA P695 2009). Selection of archetypical
building configurations also depends on the objective and restrictions posed by the nature
of the study (Gaetani d’Aragona et al. 2019; Kircher et al. 2010). Based on the architectural
and structural drawings of several real buildings in India, the authors have selected a
benchmark space frame having 9 × 3 bays as the archetype configuration (Fig. 3a). The list
of key design variables with different values considered for each of them is as follows:
 JOURNAL OF EARTHQUAKE ENGINEERING 7

Figure 3. (a) Plan and (b) section details with required longitudinal reinforcement for the example
building. All beams are 400 × 750 mm in size.

● Ductility – Buildings having high-ductility (IS 13920 1993).

● Seismic zone – III, IV, and V (PGA for MCE of 0.16 g, 0.24 g, and 0.36 g) (IS 1893
(part 1) 2002).
● Height – Buildings between 2 and 12-story high.
● Bay-width – 8200 mm (benchmark) and 6000 mm (smaller commercial buildings,
and to observe the effect of bay-width).
● Number of Bays – 3 (benchmark) and 5 (to observe the effect of number of bays).

In order to systematically characterize these design variables and to understand their

effect, thirty archetypical building configurations were selected. Further details on arche-
typical building selection have been provided in Badal and Sinha (2019). Table 3 shows the
list of thirty bare special RC frame buildings that constitute the index archetype design.
The selected buildings have been used to estimate the performance of special RC frame
buildings complying with Indian standards for design and detailing (IS 13920 1993; IS 456
2000). The effect of depth of foundation has been incorporated by longer columns on the
ground floor compared to floor height. For the same number of stories in a building,
longer ground floor columns (4.5 m compared to 3.9 m of typical story height) has two
effects. On the one hand, the building with shorter ground floor height will require smaller
cross-section for columns due to smaller lever arm for lateral forces. This leads to
relatively flexible ground floor columns resulting in the tendency to increase the funda-
mental time period of the buildings compared to those with longer ground floor height.
On the other hand, shorter length of columns increases their stiffness and contributes to a
reduction in the fundamental time period. These two factors are counteracting and hence
the effect of change in column length at ground floor is not predictable. In the buildings
8 P. S. BADAL AND R. SINHA

Table 3. Archetypical building configuration and details.

S. Performance Bldg Seismic Bay-width Ta T1 Cs
No. Group ID zonea Stories (mm) #Bays (sec)b (sec) (%)c
1 PG-1 2207 III 7 8200 3 0.91 2.56 1.8
2 2209 III 12 8200 3 1.35 3.96 1.2
3 PG-2 2211 IV 2 8200 3 0.37 1.03 6.0
4 2213 IV 4 8200 3 0.61 1.68 4.0
5 2451 IV 5 8200 3 0.71 2.00 3.4
6 2453 IV 6 8200 3 0.81 2.26 3.0
7 2215 IV 7 8200 3 0.91 2.48 2.6
8 2217 IV 12 8200 3 1.35 3.72 1.8
9 PG-3 2219 V 2 8200 3 0.37 0.93 9.0
10 2221 V 4 8200 3 0.61 1.50 5.9
11 2433 V 5 8200 3 0.71 1.78 5.1
12 2435 V 6 8200 3 0.81 2.03 4.4
13 2223 V 7 8200 3 0.91 2.32 4.0
14 2457 V 8 8200 3 1.00 2.66 3.6
15 2459 V 9 8200 3 1.10 2.94 3.3
16 2461 V 10 8200 3 1.18 3.25 3.0
17 2463 V 11 8200 3 1.27 3.55 2.8
18 2225 V 12 8200 3 1.35 3.65 2.7
19 PG-4 2227 V 4 8200 5 0.61 1.49 5.9
20 2437 V 5 8200 5 0.71 1.79 5.1
21 2439 V 6 8200 5 0.81 2.04 4.4
22 2229 V 7 8200 5 0.91 2.32 4.0
23 PG-5 2231 V 4 6000 3 0.61 1.64 5.9
24 2441 V 5 6000 3 0.71 1.92 5.1
25 2443 V 6 6000 3 0.81 2.21 4.4
26 2233 V 7 6000 3 0.91 2.55 4.0
27 PG-6 2235 V 4 6000 5 0.61 1.65 5.9
28 2445 V 5 6000 5 0.71 1.91 5.1
29 2447 V 6 6000 5 0.81 2.21 4.4
30 2237 V 7 6000 5 0.91 2.54 4.0
a
Zone-III, IV, and V correspond to PGA values at MCE of 0.16 g, 0.24 g, and 0.36 g, respectively.
b
IS 1893 (part 1) (2002) uses empirical time period expression of 0.075 × H0.75 sec. for RC bare frames.
c
Design base shear coefficient expressed as a fraction of the seismic weight of the building.

considered in the present study, they are not found to make a significant difference to the
time period of the building.

3.2. Design and Detailing of Archetypical Buildings

The structural design of the special RC frame archetype buildings was carried out by
following the RC building design standard (IS 456 2000), seismic design standard (IS 1893
(part 1) 2002), and ductile detailing standard (IS 13920 1993). The buildings have been
designed as per the design procedures prevalent in the industry. Design for regular
buildings is carried out using an equivalent lateral force (ELF) method. Base shear as
per ELF method is determined as follows:
Vb ¼ ðZ=2Þ ðI=RÞ ðSa =g Þ We (5)
where Z is the zone factor equal to the PGA corresponding to MCE; I is the important
factor of the building (= 1.0 for the ordinary buildings); Sa/g is the spectral ordinate (a
function of soil strata and building’s fundamental period and damping); and We is the
seismic weight of the building.
 JOURNAL OF EARTHQUAKE ENGINEERING 9

The empirical building fundamental period, denoted by Ta, as specified in Indian

standards and design base shear coefficients for archetype buildings are given in
Table 3. Design base shear value ranges from 1.2% to 9.0% of seismic weight for the
index archetype special RC frame buildings. Except for 10-storied and taller arche-
typical buildings, M40 (fck = 40 MPa) grade concrete has been used for both columns
and beams. For taller buildings, M50 (fck = 50 MPa) grade concrete has been used for
columns. Definition of fck is based on the characteristic strength of 150 mm cube. A
factor of 0.80 is used for converting cube strength fck to cylinder strength fc'.
Reinforcement steel of grade Fe500 having 0.2% proof stress of 500 MPa has been
used. More details on loads, design, and detailing for the buildings are provided in
Table 4. Three-dimensional linear elastic models were used for the design of the
buildings. Figure 3 shows the details of an example building (Building ID-2221). The
column cross-section has been reduced after every two or three floors except in the
buildings up to four stories, where all floors have the same column cross-section.
Further details on the selection and design of the archetypical buildings are provided
in Badal and Sinha (2019).

4. Building Modeling and Ground Motion Selection

4.1. Nonlinear Structural Model
For seismic assessment, the present study employs incremental dynamic analyses of a
set of 30 archetypical special RC frame buildings selected as described in 3. Two-
dimensional analytical models using open-source simulation software OpenSees

Table 4. Loads, design, and detailing of archetype special RC frame buildings.

S.
No. Particulars Description
1 Frame type Special moment resisting bare space frame
2 Earthquake design lateral Equivalent lateral force method
force method
3 Loading Slab thickness- 250 mm for 8.2 m span; 200 mm for 6.0 m span;
Superimposed loada (Partition, services, floor finish)- 2 kN/m2 (IS 875 (part 1) 1987);
Imposed load- 4 kN/m2 (IS 875 (part 2) 1987) Office buildings common area
4 Limit state load (1.5 DL + 1.5 IL)
combinations (1.5 DL + 1.5 EQL)
(1.2 DL + 1.2 IL + 1.2 EQL)
5 Stepping of sections Sizes of the columns are reduced after 3 or 4 floors for the buildings higher than 4
stories. IS code does not impose any strong column-weak beam criteria. Besides,
reduction in column size is a common design practice in India
6 Accidental eccentricity 5% of plan dimension perpendicular to the direction of seismic force (IS 1893 (part 1)
2002)
7 General ductile detailing 135° hooks for stirrups with 10 × diameter of extension; shear capacity more than
shear corresponding to 1.4 times the unfactored moment capacity; additional 10 ×
diameter for anchoring of reinforcements; transverse reinforcement diameter- min. 8
mm (IS 13920 1993)
8 Beam ductile detailing Spacing of hoops less than d/4 over a length of 2d from ends and d/2 elsewhere (IS
13920 1993)
9 Column ductile detailing Special confinement zone at the column ends for a length of max(cross-section
dimension, clear span/6, 450 mm); for rectangular hoops, confinement reinforcement
area ratio of 0.18 (fck/fy)(Ag/Ak−1.0)
10 Joint ductile detailing Same as that in special confinement zone at the column ends except for interior
joints
11 SCWB ratio Indian standards do not have any provisions for SCWB ratio
12 Irregularities None of the archetype buildings have any kind of plan, vertical, or torsional
irregularities
a
Superimposed load is treated as a dead load for load combination.
10 P. S. BADAL AND R. SINHA

(McKenna, Fenves, and Scott 2000) have been developed with flexural nonlinearity
modeled as concentrated plasticity at the ends of each element. For moment-resisting
frames with orthogonal lateral load-resisting systems, modeling of axial force and in-
plane bending is sufficient to capture the nonlinear behavior of the building (Haselton
et al. 2011; Liel, Haselton, and Deierlein 2011). This approach has been extensively
used in the past to assess the seismic performance of buildings where effect of torsion
is not significant (Haselton et al. 2011; Liel, Haselton, and Deierlein 2011; Ramirez et
al. 2012; Tesfamariam, Sánchez-Silva, and Rajeev 2013) and has also been adopted in
the present study. Various other researchers have shown the adequacy of concentrated
plasticity models calibrated to test data for capturing the nonlinearity up to collapse
(Haselton and Deierlein 2006; Ibarra and Krawinkler 2005). Figure 4 schematically
shows a sub-assemblage of the assessment model. Geometric nonlinearity (P-Δ effect)
has been captured using a leaning column frame.
Capacity-based shear design of columns and beams of special RC frame buildings as
specified in IS 13920 (1993) is more stringent than similar guidelines in ACI 318 (2014)
and EC 8 (2004). Several full-scale experimental test results of RC frame members
designed for capacity-based shear force as per ACI 318 (1999, 2014) confirm that flexure
yielding is the dominant failure mode (Ebrahimian et al. 2018; Jeong and Elnashai 2004).
Relative values of nominal shear strength Vn computed without using a partial safety
factor of material and plastic shear demand Vp corresponding to the plastic hinge
formation indicate the mode of failure. Detailed calibration studies in the published

Figure 4. Schematics diagram of a sub-assemblage of the nonlinear assessment model for special RC
frame buildings.
 JOURNAL OF EARTHQUAKE ENGINEERING 11

literature based on compiled experimental data reveal that shear failure modeling is not
required when the ratio Vn =Vp  1.4 (ASCE 41 2006; ATC 78-1 2012). In the present
study, it is found that Vn =Vp is more than 2.3 for columns and more than 1.4 for beams of
all 30 archetypical buildings. Therefore, in order to reduce the degree of nonlinearity in
the analytical model, the modeling of shear nonlinearity has not been considered during
numerical simulations. Shear non-linearity can be very crucial for non-code conforming buildings
Record-to-record variability, βRTR , has been estimated by analyzing the buildings for
different time history records. Effects of non-simulated modes of failure have been
approximated through the appropriate uncertainty values (Liel, Haselton, and Deierlein
2011). Total system uncertainty is given as:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
βTOT ¼ ðβ2RTR þ β2DR þ β2TD þ β2MDL Þ (6)

where DR, TD, and MDL represent design requirement, test data, and modeling, respec-
tively. FEMA P695 (2009) recommendations have been used to select these uncertainty
components.
The member model in the present paper uses Ibarra-Medina-Krawinkler (IMK) hys-
teretic rules for beam-columns (Ibarra, Medina, and Krawinkler 2005). The IMK model
captures strength and stiffness degradation, which are particularly crucial for simulating
the response to long-duration time history records. Key points on the backbone curve
have been derived using semi-empirical expressions proposed in the published literature
(Haselton et al. 2008, 2016; Panagiotakos and Fardis 2001). Specifically, secant stiffness to
the 40% of yield moment, EIstf40 has been used for pre-yield stiffness. Equation (7)
expresses EIstf40 as follows (Haselton et al. 2008):
EIstf 40 Ls EIstf 40
¼ 0:02 þ 0:98ν þ 0:09 ; where 0:35  < 0:80; (7)
EIg H EIg
where ν ¼ P=Ag fc 0 is the initial axial load ratio due to expected gravity loads (1.05 × dead
load + 0.25 × live load) and Ls =H is the shear span ratio of the member. Similarly, the
ratio of capping moment to yield moment capacity is given as:
Mc
¼ 1:25ð0:89Þν ð0:91Þ0:01fc0 : (8)
My
Other parameters of the backbone curve, plastic rotation capacity θp and post-capping
rotation θpc are based on axial load ratio and transverse steel ratio. Table 5 shows the
typical modeling parameters for beams and columns of the archetypical buildings. Bond-
slip of longitudinal reinforcement in structural members can contribute a significant
component (approximately, one-third of θp ) to the nonlinear deformation of reinforced
concrete members (Panagiotakos and Fardis 2001; Paulay and Priestly 1992). In the
present study, bond-slip has been considered in the estimation of θp with the use of
semi-empirical relations (Haselton et al. 2016).
The beam-column joints are modeled with a diagonal compression strut mechanism.
This model matches well with the experimentally observed cyclic response of the joints
(Lowes, Mitra, and Altoontash 2003). The joint shear panel stiffness is based on Meinheit
and Jirsa (1981). The expected compressive strength of concrete is fck,exp = fck + 1.645σ,
where standard deviation, σ, has been taken as 5 MPa as per Indian standards (IS 456
 12

Table 5. Details of central column and beam at ground floor level for archetypical buildings.
Seismic Section Long. reinf. Trans. reinf. θcap = θpc
Bldg. ID zone Stories Member fck (MPa) (mm2) ρl ρsh Axial load ratio, \nu Mc/ My θp+θy (rad)
2207 III 7 Column 40 600 × 600 3.03% 0.52% 0.33 1.15 0.034 0.040
Beam 40 350 × 750 0.93% 0.38% 0.00 1.21 0.043 0.100
2209 III 12 Column 50 700 × 700 3.17% 0.45% 0.35 1.14 0.030 0.032
Beam 40 350 × 750 0.92% 0.38% 0.00 1.21 0.043 0.100
2211 IV 2 Column 40 450 × 450 2.26% 0.70% 0.17 1.18 0.045 0.100
Beam 40 350 × 750 0.90% 0.46% 0.00 1.21 0.058 0.100
2213 IV 4 Column 40 500 × 500 3.58% 0.63% 0.27 1.16 0.043 0.061
Beam 40 350 × 750 1.16% 0.46% 0.00 1.21 0.053 0.100
2451 IV 5 Column 40 500 × 500 4.14% 0.63% 0.34 1.15 0.038 0.046
P. S. BADAL AND R. SINHA

Beam 40 350 × 750 1.09% 0.46% 0.00 1.21 0.052 0.100

2453 IV 6 Column 40 550 × 550 3.43% 0.57% 0.34 1.15 0.036 0.042
Beam 40 350 × 750 1.09% 0.46% 0.00 1.21 0.052 0.100
2215 IV 7 Column 40 600 × 600 2.88% 0.52% 0.34 1.15 0.034 0.039
Beam 40 350 × 750 1.16% 0.46% 0.00 1.21 0.053 0.100
2217 IV 12 Column 50 700 × 700 3.03% 0.45% 0.36 1.14 0.030 0.031
Beam 40 400 × 750 1.01% 0.40% 0.00 1.21 0.050 0.100
2219 V 2 Column 40 500 × 500 2.58% 0.63% 0.14 1.18 0.051 0.100
Beam 40 350 × 750 1.05% 0.46% 0.00 1.21 0.059 0.100
2221 V 4 Column 40 550 × 550 3.43% 0.57% 0.23 1.17 0.046 0.067
Beam 40 400 × 750 1.24% 0.52% 0.00 1.21 0.057 0.100
2433 V 5 Column 40 550 × 550 3.84% 0.57% 0.29 1.16 0.041 0.053
Beam 40 400 × 750 1.18% 0.42% 0.00 1.21 0.046 0.100
2435 V 6 Column 40 600 × 600 3.23% 0.52% 0.29 1.16 0.038 0.048
Beam 40 400 × 750 1.22% 0.42% 0.00 1.21 0.046 0.100
2223 V 7 Column 40 600 × 600 3.92% 0.52% 0.34 1.15 0.035 0.039
Beam 40 400 × 750 1.28% 0.50% 0.00 1.21 0.062 0.100
2457 V 8 Column 40 650 × 650 3.51% 0.48% 0.33 1.15 0.034 0.037
Beam 40 400 × 750 1.28% 0.52% 0.00 1.21 0.057 0.100
2459 V 9 Column 40 650 × 650 3.81% 0.48% 0.37 1.14 0.031 0.031
Beam 40 400 × 750 1.45% 0.45% 0.00 1.21 0.058 0.100
2461 V 10 Column 50 650 × 650 3.21% 0.48% 0.34 1.14 0.032 0.035
Beam 40 400 × 750 1.30% 0.45% 0.00 1.21 0.056 0.100
2463 V 11 Column 50 650 × 650 3.81% 0.48% 0.38 1.13 0.030 0.030
Beam 40 400 × 750 1.34% 0.45% 0.00 1.21 0.057 0.100
2225 V 12 Column 50 700 × 700 3.28% 0.45% 0.36 1.14 0.030 0.031
Beam 40 400 × 750 1.34% 0.45% 0.00 1.21 0.057 0.100
2227 V 4 Column 40 550 × 550 3.25% 0.57% 0.23 1.17 0.042 0.068
Beam 40 400 × 750 1.19% 0.52% 0.00 1.21 0.057 0.100
2437 V 5 Column 40 550 × 550 3.84% 0.57% 0.29 1.16 0.041 0.053
Beam 40 400 × 750 1.18% 0.42% 0.00 1.21 0.046 0.100
2439 V 6 Column 40 600 × 600 3.23% 0.52% 0.29 1.16 0.038 0.048
Beam 40 400 × 750 1.19% 0.42% 0.00 1.21 0.052 0.100
2229 V 7 Column 40 600 × 600 3.92% 0.52% 0.34 1.15 0.035 0.039
Beam 40 400 × 750 1.28% 0.50% 0.00 1.21 0.062 0.100
2231 V 4 Column 40 400 × 400 4.03% 0.79% 0.20 1.17 0.052 0.100
Beam 40 300 × 600 0.99% 0.40% 0.00 1.21 0.049 0.100
2441 V 5 Column 40 450 × 450 2.83% 0.70% 0.20 1.17 0.048 0.092
Beam 40 300 × 600 1.15% 0.40% 0.00 1.21 0.050 0.100
2443 V 6 Column 40 450 × 450 3.45% 0.70% 0.24 1.16 0.049 0.077
Beam 40 300 × 600 1.09% 0.40% 0.00 1.21 0.050 0.100
2233 V 7 Column 40 450 × 450 3.88% 0.70% 0.28 1.16 0.041 0.065
Beam 40 300 × 600 1.17% 0.40% 0.00 1.21 0.050 0.100
2235 V 4 Column 40 400 × 400 4.03% 0.79% 0.20 1.17 0.052 0.100
Beam 40 300 × 600 1.07% 0.54% 0.00 1.21 0.055 0.100
2445 V 5 Column 40 450 × 450 2.91% 0.93% 0.20 1.17 0.058 0.100
Beam 40 300 × 600 1.09% 0.40% 0.00 1.21 0.050 0.100
2447 V 6 Column 40 450 × 450 3.18% 0.70% 0.24 1.16 0.044 0.078
Beam 40 300 × 600 1.09% 0.84% 0.00 1.21 0.065 0.100
2237 V 7 Column 40 450 × 450 3.88% 0.70% 0.28 1.16 0.041 0.065
Beam 40 300 × 600 1.17% 0.40% 0.00 1.21 0.050 0.100
JOURNAL OF EARTHQUAKE ENGINEERING
13
14 P. S. BADAL AND R. SINHA

2000). For steel, in the absence of any such distribution prescribed by Indian standards,
the expected strength has been taken as 1.2 times the yield strength (PEER/ATC 72-1
2010) assuming that steel manufacturing process in the region is closely aligned with
global practices.
For nonlinear time history response, accurate modeling of damping is crucial (Hall
2006; Petrini et al. 2008; Jehel, Léger, and Ibrahimbegovic 2014). Since the empirical
relations used for the key points on the backbone curve already consider energy dissipa-
tion due to damping in the post-yield range, nonlinear springs having high stiffness values
have been modeled without damping. For a pre-yield range of the response of the
members, a Rayleigh damping of 5% in first and third mode has been employed. In
order to check the capability of developed nonlinear model to predict seismic behavior
and peak responses, the model has been validated with experimental results of a ductile
reinforced concrete frame tested by Filiatrault, Lachapelle, and Lamontagne (1998).
Details of the validation of numerical model has been presented in Appendix A.

4.2. Selection of Ground Motions

For nonlinear time history response analysis, Jayaram et al. (2011) showed that it is crucial
for time history records to represent mean as well as the variance of the target response
spectrum. Furthermore, they found FEMA P695 (2009) ground motion records to meet
these requirements. The present study uses FEMA P695 (2009) 22 × 2 far-field records.
Figure 5 depicts the response spectra of these ground motion records along with 5 and 95
percentile spectra. To demonstrate the agreement of median response spectrum of selected
ground motion suite, the design response spectrum from IS 1893 (part 1) (2002) has been
scaled at the time period of 1.0 sec that lies well within 0.37–1.35 sec, the range of time
period of archetypical buildings. Scaling of ground motion records for collapse analysis
during IDA is carried out as per guidelines of FEMA P695 (2009). Since the current study
focusses on the buildings that may be located anywhere in India, adjustment through the

Figure 5. Response spectra of 22 × 2 suite of FEMA P695 far-field ground motion. Blue solid line represents
IS 1893 (part 1) (2002) response spectrum scaled to match median spectral acceleration at 1 sec.
 JOURNAL OF EARTHQUAKE ENGINEERING 15

use of site-specific spectral shape factor has not been adopted (Haselton and Baker 2006;
FEMA P695 2009).

5. Analysis Results
5.1. Observed OSF for Special RC Frame Buildings
Figure 6 shows the result of a pushover analysis of sample building (Building ID-2221).
The figure shows maximum normalized base shear, Cs = 17.48%, and the design base shear
coefficient, Cs = 5.95%. As defined in 2.2, the OSF value, Ω0 , for this building is found to
be 17.48/5.95 = 2.94. Table 6 presents the values of Cs,max and OSF for all archetypical
buildings considered in this study. The OSF is found to be generally more than 2.50. The
corresponding OSF value in ASCE 7 (2010) is 3.0. These results show that the ductile RC
frame buildings complying with Indian standards are associated with smaller inherent
strength than the buildings with the same lateral load-resisting system and complying with
international standards.
It can be observed that there is an increase in Ω0 as seismicity reduces. For instance, the
seven-story building of PG-3 (Building ID-2223) in seismic zone-V has an Ω0 value of 2.77
and 7-story buildings in PG-2 and PG-1, i.e., buildings in seismic zone-IV and III, have
the value of Ω0 as 3.36 and 4.40, respectively. This is attributed to the non-seismic load
combinations becoming critical for structural members in low or moderate seismic
regions (seismic zone-III and IV). Further, with a lower limit of 2.50, taller buildings
are associated with smaller values of Ω0 . Based on the results presented in Table 6, the
following expression for the OSF of special RC frame buildings as a function of seismic
zone factor, Z and code-specified time period, Ta is proposed.

Ω0 ðZ; Ta Þ ¼ 2:50  f1 ðZÞ  f2 ðTa Þ  2:50; (9)

Figure 6. Pushover curve for a sample building (building ID-2221). Coefficients Cs;max and Cs correspond
to maximum base shear capacity and design base shear, respectively.
16 P. S. BADAL AND R. SINHA

Table 6. Overstrength factor, Ω0 , of archetypical special RC frame buildings.

S. No. Performance Group Bldg ID Seismic zone Stories Cs,max (%) Ω0
1 PG-1 2207 III 7 7.7 4.40
2 2209 III 12 3.8 3.24
3 PG-2 2211 IV 2 20.6 3.44
4 2213 IV 4 14.0 3.54
5 2451 IV 5 11.6 3.44
6 2453 IV 6 9.9 3.36
7 2215 IV 7 8.9 3.36
8 2217 IV 12 4.7 2.68
9 PG-3 2219 V 2 27.3 3.04
10 2221 V 4 17.5 2.94
11 2433 V 5 14.7 2.91
12 2435 V 6 12.9 2.92
13 2223 V 7 11.0 2.77
14 2457 V 8 9.6 2.69
15 2459 V 9 8.6 2.62
16 2461 V 10 7.8 2.56
17 2463 V 11 7.3 2.58
18 2225 V 12 6.5 2.45
19 PG-4 2227 V 4 16.3 2.74
20 2437 V 5 14.6 2.89
21 2439 V 6 12.7 2.87
22 2229 V 7 11.0 2.78
23 PG-5 2231 V 4 15.4 2.58
24 2441 V 5 13.2 2.60
25 2443 V 6 11.2 2.52
26 2233 V 7 9.6 2.44
27 PG-6 2235 V 4 15.2 2.56
28 2445 V 5 12.8 2.53
29 2447 V 6 11.1 2.51
30 2237 V 7 9.8 2.48

where f1 ðZÞ ¼ ð0:36=ZÞ0:65  1:0; and

f2 ðTa Þ ¼ minð1=Ta ; 1:0Þ:

Here, f1 ðZÞ is the modification factor for buildings in low seismicity and f2 ðTa Þ is the
reduction factor for long-period buildings. The observed and predicted values of Ω0 are
shown in Fig. 7. As a number of archetypical buildings have the same empirical time
period, the data-points having the same empirical time period have been staggered along
the horizontal axis and are shown under a single gray band. It is noted that predicted
values are always lower bound of the observed values.
The results of pushover analyses are based on the inverse parabolic load pattern as
specified in IS 1893 (part 1) (2002). Effect of loading profile on results of pushover results
has been separately investigated, but not reproduced here in detail for the lack of space.
Overstrength factor has been found to change in the range of only 6%-8% when pushover
loading profile follows the first fundamental mode shape or loading profile as per ASCE 7
(2010). Further, due to highly dominant first mode for all archetypical buildings (in order
of 80% modal participation), the effects of higher modes are minimal.
 JOURNAL OF EARTHQUAKE ENGINEERING 17

5.0
Analytical
4.5 Predicted

4.0

Overstrength Factor, Ω0
3.5

3.0

2.5

2.0

1.5
0.37 0.61 0.71 0.81 0.91 1.00 1.10 1.18 1.27 1.35
1.0
Time Period (sec)

Figure 7. Analytical and predicted overstrength factor, Ω0 , for all archetypical buildings.

5.2. Observations from Incremental Dynamic Analysis

Figure 8 presents a set of IDA curves for a sample building while subjected to 22 pairs of
far-field ground motion records provided in FEMA P695 (2009) as discussed in 2.2.
Geometric mean of spectral acceleration along two horizontal direction, SageoMðTa Þ,
where Ta is an empirical time period as per design standard, has been considered as the
intensity measure. For each ground motion record, given a damage state ds, scaling
required for the building to reach the corresponding value of MIDR is calculated.
Median μSa;ds and standard deviation βRTR;ds fragility parameters for the building corre-
sponding to damage state ds are determined from these scaled values of SaðTa Þ.
Lognormal curve shown vertically at 4% abscissae in Fig. 8 corresponds to CP damage

Figure 8. IDA curves for a sample building (building ID-2221). Median fragility parameter, μSa;CP
corresponding to “collapse prevention” damage state (4% MIDR) is also shown by the horizontal
broken line.
18 P. S. BADAL AND R. SINHA

state. Corresponding median parameter μSa;CP is also marked in the figure. It can be noted
that due to lognormal assumption on the fragility function median values are skewed from
the peak of the probability distribution function.
It is observed from the figure that intensity measure values have a higher level of
dispersion for severe damage states, i.e., for high MIDR values. This observation can be
attributed to the increased scatter in the seismic behavior of RC members in the highly
nonlinear range. The scatter in intensity measure value also indicates that the chosen
ground motion suite duly captures the variability in the seismic performance of the
building. Similar IDA curves are observed for all the archetypical buildings.
Figure 9 shows μSa;CP values for all baseline buildings (PG-1, PG-2, and PG-3). It is
observed that the median capacity of buildings in the range of four to seven stories of
height remains roughly unchanged from seismic zone-IV to zone-V. Though design base
shear increases from zone-IV to zone-V, due to a reduction in the analytical time period
of the buildings, available nonlinear capacity is smaller than the ductility demand. This is
equivalent to the inelastic response spectrum approaching from ‘equal displacement range’
to ‘equal energy range’ (Paulay and Priestly 1992).

5.3. Observed RRF Value Based on P[CP|MCE]

Based on IDA results and performance objective implied in IS 1893 (part 1) (2002), Table
7 shows the median capacity, μSa;CP , and total uncertainty, βTOT , in the seismic fragility of
all archetypical buildings. The table also shows the probability of exceeding the perfor-
mance objective CP conditional on MCE, P[CP|MCE] and available response reduction
factor, Ravl, for all buildings. Buildings in seismic zone-III invariably have a very low
probability of exceeding the target performance objective. Design of vertical members in
the low-seismic zone is generally governed by gravity loads, and thus, they tend to have
high reserve strength against lateral forces. All building configurations in seismic zone-IV,
except the two-storied building which fails by a narrow margin (10.2% P[CP|MCE] value
against a desirable value of 10% or less), are found to meet the performance objective.

1.8
Seismic Zone-III
Seismic Zone-IV
Seismic Zone-V
Median CP capacity, µSa,CP (g)

1.5

1.2

0.9

0.6
0 2 4 6 8 10 12 14
Number of Stories

Figure 9. Median collapse prevention capacity of all baseline buildings in PG-1, PG-2, and PG-3,
designed for seismic zone-III, IV, and V, respectively.
 JOURNAL OF EARTHQUAKE ENGINEERING 19

Table 7. Response reduction factor for all archetypical special RC frame buildings.
S. No. Bldg. ID Seismic Zone Stories μSa,CP (g) βTOT P[CP|MCE] (%) Ravl
1 2207 III 7 0.93 0.59 0.2 12.4
2 2209 III 12 0.66 0.68 0.6 11.7
3 2211 IV 2 1.36 0.64 10.2 5.0
4 2213 IV 4 1.13 0.65 5.4 6.2
5 2451 IV 5 1.11 0.61 2.5 7.5
6 2453 IV 6 0.96 0.56 1.8 7.9
7 2215 IV 7 1.00 0.60 1.3 8.8
8 2217 IV 12 0.72 0.68 2.0 8.5
9 2219 V 2 1.56 0.58 17.3 4.1
10 2221 V 4 1.12 0.64 16.2 4.1
11 2433 V 5 1.15 0.56 7.1 5.5
12 2435 V 6 0.99 0.51 5.9 5.8
13 2223 V 7 1.03 0.55 4.1 6.4
14 2457 V 8 0.89 0.55 5.1 6.1
15 2459 V 9 0.74 0.56 7.4 5.5
16 2461 V 10 0.74 0.62 7.7 5.5
17 2463 V 11 0.74 0.63 6.5 5.8
18 2225 V 12 0.76 0.63 4.9 6.3
19 2227 V 4 1.01 0.61 19.4 3.9
20 2437 V 5 1.08 0.57 9.1 5.2
21 2439 V 6 0.94 0.51 7.1 5.5
22 2229 V 7 0.93 0.50 4.2 6.2
23 2231 V 4 1.10 0.60 15.2 4.3
24 2441 V 5 0.94 0.50 10.5 4.9
25 2443 V 6 0.97 0.54 7.1 5.5
26 2233 V 7 0.83 0.55 8.8 5.2
27 2235 V 4 1.06 0.59 16.6 4.2
28 2445 V 5 0.88 0.49 13.4 4.6
29 2447 V 6 0.87 0.52 9.3 5.1
30 2237 V 7 0.82 0.52 8.3 5.3

These results show that when the performance assessment of RC frame buildings in
seismic zone-IV is carried out under probabilistic framework, they are found to meet
the implicitly defined performance objective of life safety with the code-specified RRF
value, unlike the results from simpler analysis presented by Mondal, Ghosh, and Reddy
(2013). However, the Table also shows that several buildings located in seismic zone-V
exceed 10% value of P[CP|MCE]. It can be observed that these buildings may have nearly
twice the probability of exceeding collapse prevention performance than the commonly
accepted threshold (the probability is 19.4% for Building ID-2227). This shows that the
design of these building configurations as per present design standards is not sufficient to
meet the expected performance objective.
Figure 10 shows the variation of the available RRF value, Ravl against the height
of the building for performance groups PG-3, PG-4, PG-5, and PG-6, i.e., for all
buildings located in seismic zone-V. Across all performance groups, Ravl generally
increases with height. Furthermore, low-and mid-rise buildings up to five-story in
seismic zone-V are found to be deficient (having Ravl value of less than 5.0). It is
also noted from Fig. 10 that curves of Ravl values for these performance groups
intercross one another. This indicates that bay-width and number of bays do not
impact the RRF value significantly, confirming that for moment-frame buildings,
plan dimensions do not significantly influence the building seismic performance.
20 P. S. BADAL AND R. SINHA

Building Height (m)

16.2 20.1 24.0 27.9
7

Response Reduction Factor (RRF)

6

RRF as per IS 1893

(part 1) (2002)
5

4
3 bays
5 bays
8.2 m bay-width
6.0 m bay-width
3
4 5 6 7
Number of Stories

Figure 10. Effect of number of bays, bay-width, and number of stories on RRF value for buildings in
seismic zone-V (PG-3, PG-4: 8.2 m bay-width with 3 and 5 bays, respectively; PG-5, PG-6: 6.0 m bay-
width with 3 and 5 bays, respectively).

5.4. Observed RRF Value Based on λCP and CPMR

Though zone factor values assigned to different seismic zones serve the purpose of PGA
corresponding to MCE, IS 1893 (part 1) (2002) does not distinctly define MCE-level ground
motion. Therefore, available RRF value Ravl evaluated as per (4), exclusively based on MCE-level
seismic hazard, deserves further consideration. In this regard, two additional seismic perfor-
mance metrics were chosen to ascertain the robustness of Ravl values presented earlier in 5.3.
First, annual probability of exceeding collapse prevention, λCP considers all possible ground
motion levels and is obtained by convoluting the seismic hazard curve with fragility function
corresponding to CP damage state. Due to an absence of rigorous peer-reviewed probabilistic
seismic hazard map for the region of interest, hazard curve has been approximated using the
frequently adopted two-parameter model in terms of power function given as
PðIM > imÞ, k0 imk (Cordova et al. 2000; Jalayer and Cornell 2003; EC 8 2004). Second
metric, collapse prevention margin ratio, CPMR is the ratio of median fragility parameter
corresponding to CP to SMT , as defined earlier in 2.2. Table 8 presents the observed and
acceptable CPMR values along with the observed λCP values. Acceptable value of λCP is
considered as 2 × 10−4 (Luco et al. 2007; ASCE 7 2010). The table also indicates whether a
building is deficient on these criteria. Acceptable values for CPMR are based on total system
uncertainty, βTOT . Adopted from FEMA P695 (2009), these values are properties of statistical
distribution and hence, they continue to remain valid irrespective of the target performance level.
Interestingly, the deficient archetypical buildings identified from the criterion of P[CP|
MCE] in 5.3, are also found to fall short of meeting the acceptable values of these two
criteria, i.e., λCP and CPMR. For instance, building ID-2227 was found to have a value of
λCP of 3.3 × 10−4 that exceeds the acceptable limit by 65%. The only exceptions are
building ID-2211 and 2441, having 10.2% and 10.5% value of P[CP|MCE] against the
threshold of 10%, which are very close to the acceptable limit and were found to be
 JOURNAL OF EARTHQUAKE ENGINEERING 21

Table 8. Comparison of values of different seismic performance metrics for archetypical special RC
frame buildings.
S. Bldg. P[CP|MCE] Based on P[CP| Observed Acceptable Based on λCP Based on
No. ID (%) MCE] CPMR10% CPMR10% CPMR (10−4) λCP
1 2207 0.2 OK 5.31 2.13 OK 0.2 OK
2 2209 0.6 OK 5.62 2.39 OK 0.2 OK
3 2211 10.2 Not-OK 2.27 2.28 Not-OK 1.9 OK
4 2213 5.4 OK 2.85 2.30 OK 1.1 OK
5 2451 2.5 OK 3.30 2.19 OK 0.7 OK
6 2453 1.8 OK 3.24 2.05 OK 0.6 OK
7 2215 1.3 OK 3.79 2.16 OK 0.5 OK
8 2217 2.0 OK 4.08 2.41 OK 0.5 OK
9 2219 17.3 Not-OK 1.73 2.11 Not-OK 2.9 Not-OK
10 2221 16.2 Not-OK 1.88 2.28 Not-OK 2.8 Not-OK
11 2433 7.1 OK 2.28 2.06 OK 1.4 OK
12 2435 5.9 OK 2.23 1.93 OK 1.3 OK
13 2223 4.1 OK 2.61 2.03 OK 1.0 OK
14 2457 5.1 OK 2.48 2.03 OK 1.1 OK
15 2459 7.4 OK 2.26 2.06 OK 1.4 OK
16 2461 7.7 OK 2.42 2.22 OK 1.5 OK
17 2463 6.5 OK 2.60 2.25 OK 1.3 OK
18 2225 4.9 OK 2.85 2.25 OK 1.0 OK
19 2227 19.4 Not-OK 1.69 2.19 Not-OK 3.3 Not-OK
20 2437 9.1 OK 2.14 2.08 OK 1.7 OK
21 2439 7.1 OK 2.12 1.93 OK 1.4 OK
22 2229 4.2 OK 2.36 1.89 OK 1.1 OK
23 2231 15.2 Not-OK 1.85 2.15 Not-OK 2.5 Not-OK
24 2441 10.5 Not-OK 1.86 1.89 Not-OK 1.9 OK
25 2443 7.1 OK 2.20 1.99 OK 1.4 OK
26 2233 8.8 OK 2.09 2.01 OK 1.6 OK
27 2235 16.6 Not-OK 1.77 2.14 Not-OK 2.7 Not-OK
28 2445 13.4 Not-OK 1.73 1.89 Not-OK 2.2 Not-OK
29 2447 9.3 OK 1.97 1.94 OK 1.7 OK
30 2237 8.3 OK 2.06 1.95 OK 1.6 OK

deficient based on the values of P[CP|MCE]. These observations further strengthen the
following proposed modifications in the seismic design parameters.

5.5. Recommended RRF Values for Special RC Moment-Frame Buildings

It can be seen from the above discussion that Indian seismic design standards overestimate
the RRF value for the low- and mid-rise special RC frame buildings located in seismic
zone-V. Thus, these buildings should be designed with a smaller RRF than that specified
by design code (a fixed value of 5.0), i.e., for higher design base shear. Based on the
present study, it has been found that for structures in seismic zone-V with the empirical
time period of less than 0.6 sec, adoption of RRF value of 4.0 can meet the seismic
performance objective. It is also seen that for buildings with an empirical time period
longer than 0.9 sec, RRF value of 5.0 meets seismic performance objective. In the present
study, a linear variation of RRF value within this range is proposed. Accordingly, the
proposed expression for RRF in seismic zone-V denoted by R0 is as follows:
8
<4 ; 0  Ta  0:60
R0 ¼ 2 þ 10 3 a ; 0:60 < Ta  0:90
T (10)
:
5 ; Ta > 0:90;
22 P. S. BADAL AND R. SINHA

Figure 11. Proposed RRF, R0, for the design of special RC frame buildings located in seismic zone-V.

where Ta is the code-specified empirical time period of the buildings in seconds. Figure 11 shows
R0 along with analytically determined RRF value, Ravl . To keep the expression simple and easy to
adopt, the modification has been expressed as a function of Ta. It is also noted that in other
seismic zones, the currently prescribed RRF value of 5.0 does not need to be modified.

6. Summary and Conclusions

The present paper presents a probabilistic-based framework for the assessment of seismic
performance of RC building design complying with Indian seismic standards. These
standards were drawn up before the advancement in probabilistic analytical tools and
the availability of comprehensive experimental database for calibration of member per-
formance. Seismic behavior of a set of 30 archetypical buildings, designed as per the
prevalent practices, has been assessed with respect to the implicit performance objective
envisaged in the design standards. A probabilistic framework has been adopted to con-
sider inherent uncertainty in the seismic demand and structural capacity.
It is found that several building configurations in high seismic regions, when designed as per
Indian standards, may not possess sufficient inelastic capacity to meet the expected performance
as implicitly envisaged by the seismic code. Modification to the RRF values of special RC
moment-frame buildings has been proposed to overcome deficiencies in the design of these
buildings while continuing to follow the prescriptive design procedure in the code.
The following conclusions are drawn from the present paper:

(1) A framework to incorporate the probabilistic seismic performance of RC buildings

in the prescriptive design procedure specified in Indian seismic design standards
has been proposed. It relies on rigorous nonlinear dynamic analyses of a set of
archetype buildings representing a lateral load-resisting building system and inte-
grates several uncertainties in the seismic performance of the buildings while also
considering the degradation in strength and stiffness of buildings and the influence
 JOURNAL OF EARTHQUAKE ENGINEERING 23

of dynamic characteristics of the load and buildings. It is found that probabilistic

seismic performance objectives can be achieved in a force-based prescriptive design
procedure through simple modifications to the current procedure.
(2) Ductile RC frame buildings complying with Indian standards have lower bound of
overstrength factor of 2.5. This is smaller than the overstrength factor for the same
lateral load-resisting systems complying with international standards (for instance, 3.0 as
per ASCE-7). This implies that observed vulnerability of buildings after damaging
earthquakes in parts of the world where these standards are applicable should not be
directly used to develop fragility and vulnerability functions for similar configurations of
RC buildings in India.
(3) The OSF reduces with severity in the seismicity and with an increase in the height of the
building. A generalized expression for OSF that takes into account the seismic hazard and
the building’s code-specified time period has been proposed in this paper.
(4) It is concluded from the present study that contrary to the observations of Mondal,
Ghosh, and Reddy (2013), building design for seismic zone-IV meets the implicitly
defined performance objective of life safety in Indian standards. Thus, the currently
prescribed RRF value of 5.0 is found to be adequate for buildings in seismic zone-IV.
(5) The low- and mid-rise buildings located in seismic zone-V are not able to meet the
expected seismic performance. Some of these buildings have nearly twice the probability
of exceeding (as high as 19.4%) collapse prevention performance level at MCE ground
motion than the commonly accepted threshold of 10%. Design of these building con-
figurations as per present design standards is not sufficient to meet the expected
performance objective. To overcome this deficiency, modification to RRF prescribed in
the Indian standards for special RC moment-frame buildings has been proposed.
(6) Annual probability of exceeding collapse prevention, λCP , an alternate seismic
performance metric considering all possible earthquake scenario, is found to
accurately identify the deficiency in low- and mid-rise buildings. Deficient buildings
are found to exceed the target probability by as high as 65%. Further, this frame-
work can be easily adopted to derive design parameters for special buildings with
stringent limits on the annual probability of exceedance of any damage state.
(7) It is found that the effect of the number of bays of an RC frame building on RRF is
not significant. It is also found that bay-width does not have a noticeable impact on
RRF. This behavior confirms that, for RC moment frame buildings, plan dimen-
sions do not influence their seismic performance. This study validates the present
practice to specify empirical time period expression for bare frame buildings that is
independent of the plan dimensions.

ORCID
Prakash S. Badal http://orcid.org/0000-0003-4097-8444
Ravi Sinha http://orcid.org/0000-0002-3371-4355

References
ACI 318. 1999. Building code requirements for structural concrete and commentary (ACI 318-99).
MI, USA: American Concrete Institute, Farmington Hills.
24 P. S. BADAL AND R. SINHA

ACI 318. 2014. Building code requirements for structural concrete and commentary (ACI 318-14).
MI, USA: American Concrete Institute, Farmington Hills.
ASCE 41. 2006. Seismic rehabilitation of existing buildings (ASCE/SEI 41-06). Reston, VA: American
Society of Civil Engineers.
ASCE 7. 2010. Minimum design loads for buildings and other structures (ASCE/SEI 7-10). Reston,
Virginia: American Society of Civil Engineers.
ATC 78-1. 2012. Evaluation of the methodology to select and prioritize collapse indicators in older
concrete buildings. Redwood City: Applied Technology Council, PEER.
Badal, P. S., and R. Sinha. 2019. Selection of archetypical building configuration for special reinforced
concrete moment-resisting frames. Mumbai: IIT Bombay. www.civil.iitb.ac.in/~rsinha/TechRep_
SMRF_Archetype.
Census. 2001. Census of India: Houselisting and housing census data. New Delhi, India: Office of the
Registrar General.
Census. 2011. Census of India: Houselisting and housing census data. New Delhi, India: Office of the
Registrar General.
Cordova, P. P., G. G. Deierlein, S. S. Mehanny, and C. A. Cornell. 2000. Development of a two-
parameter seismic intensity measure and probabilistic assessment procedure. In The second US-
Japan workshop on performance-based earthquake engineering methodology for reinforced concrete
building structures, ed. T. Kabeyasawa, and J. P. Moehle, 187–206. Sapporo, Hokkaido, Japan.
Denavit, M. D., J. F. Hajjar, T. Perea, and R. T. Leon. 2016. Seismic performance factors for moment
frames with steel-concrete composite columns and steel beams. Earthquake Engineering &
Structural Dynamics 45 (10): 1685–703. doi: 10.1002/eqe.2737.
Der Kiureghian, A. 2005. Non-ergodicity and PEER’s framework formula. Earthquake Engineering
& Structural Dynamics 34 (13): 1643–52. doi: 10.1002/(ISSN)1096-9845.
Ebrahimian, H., R. Astroza, J. P. Conte, and T. C. Hutchinson. 2018. Pretest nonlinear finite-
element modeling and response simulation of a full-scale 5-story reinforced concrete building
tested on the NEES-UCSD shake table. Journal of Structural Engineering 144 (3): 04018009. doi:
10.1061/(ASCE)ST.1943-541X.0001963.
EC 8. 2004. Eurocode: Design of structures for earthquake resistance. Part 1: General rules, seismic
actions and rules for buildings. The European Standard.
Ellingwood, B. R. 2001. Earthquake risk assessment of building structures. Reliability Engineering &
System Safety 74 (3): 251–62. doi: 10.1016/S0951-8320(01)00105-3.
FEMA 273. 1997. NEHRP guidelines for the seismic rehabilitation of buildings. Washington, DC:
Federal Emergency Management Agency.
FEMA 356. 2000. Prestandard and commentary for the seismic rehabilitation of buildings.
Washington, DC: Council and Building Seismic Safety, Federal Emergency Management Agency.
FEMA 445. 2006. Next-generation performance-based seismic design guidelines program plan for new
and existing buildings. Washington, DC: Federal Emergency Management Agency.
FEMA 450. 2004. NEHRP recommended provisions for seismic regulations for new buildings and
other structures, FEMA 450-2/2003 edition, part 2: Commentary. Redwood City: Federal
Emergency Management Agency.
FEMA 58-1. 2012. Seismic performance assessment of buildings. Volume 1–methodology.
Washington, DC: Federal Emergency Management Agency.
FEMA P695. 2009. Quantification of building seismic performance factors. Redwood City, California:
Applied Technology Council.
Filiatrault, A., É. Lachapelle, and P. Lamontagne. 1998. Seismic performance of ductile and
nominally ductile reinforced concrete moment resisting frames. I. Experimental study.
Canadian Journal of Civil Engineering 25 (2): 331–41. doi: 10.1139/l97-097.
Franchin, P., F. Petrini, and F. Mollaioli. 2018. Improved risk-targeted performance-based seismic
design of reinforced concrete frame structures. Earthquake Engineering & Structural Dynamics 47
(1): 49–67. doi: 10.1002/eqe.v47.1.
Gaetani d’Aragona, M., M. Polese, E. Cosenza, and A. Prota. 2019. Simplified assessment of
maximum interstory drift for RC buildings with irregular infills distribution along the height.
Bulletin of Earthquake Engineering 17: 707–36. doi: 10.1007/s10518-018-0473-y.
 JOURNAL OF EARTHQUAKE ENGINEERING 25

Hall, J. F. 2006. Problems encountered from the use (or misuse) of Rayleigh damping. Earthquake
Engineering & Structural Dynamics 35 (5): 525–45. doi: 10.1002/(ISSN)1096-9845.
Han, S. W., T. O. Kim, and S. J. Baek. 2018. Seismic performance evaluation of steel ordinary
moment frames. Earthquake Spectra 34 (1): 55–76. doi: 10.1193/011117EQS010M.
Haselton, C. B., A. B. Liel, G. G. Deierlein, B. S. Dean, and J. H. Chou. 2011. Seismic collapse safety
of reinforced concrete buildings. I: Assessment of ductile moment frames. Journal of Structural
Engineering 137 (4): 481–91. doi: 10.1061/(ASCE)ST.1943-541X.0000318.
Haselton, C. B., A. B. Liel, S. C. Taylor-Lange, and G. G. Deierlein. 2016. Calibration of model to
simulate response of reinforced concrete beam-columns to collapse. ACI Structural Journal 113
(6): 1141–52. doi: 10.14359/51689245.
Haselton, C. B., C. A. Goulet, J. Mitrani-Reiser, J. L. Beck, G. G. Deierlein, K. A. Porter, J. P.
Stewart, and E. Taciroglu. 2008. An assessment to benchmark the seismic performance of a code-
conforming reinforced-concrete moment-frame building. Pacific Earthquake Engineering
Research Center, (2007/1).
Haselton, C. B., and G. G. Deierlein. 2006. Assessing seismic collapse safety of modern reinforced
concrete moment frame buildings. Stanford, CA: Stanford University.
Haselton, C. B., and J. W. Baker. 2006. Ground motion intensity measures for collapse capacity
prediction: Choice of optimal spectral period and effect of spectral shape. In 8th U.S. National
Conference on Earthquake Engineering, San Francisco, CA, 18–22.
Ibarra, L. F., and H. Krawinkler. 2005. Global collapse of frame structures under seismic excitations.
No. TR152. Stanford, CA: Pacific Earthquake Engineering Research Center.
Ibarra, L. F., R. A. Medina, and H. Krawinkler. 2005. Hysteretic models that incorporate strength
and stiffness deterioration. Earthquake Engineering & Structural Dynamics 34 (12): 1489–512.
doi: 10.1002/eqe.495.
IS 13920. 1993. Ductile detailing of reinforced concrete structures subjected to seismic forces- code of
practice. New Delhi: Bureau of Indian Standards.
IS 1893 (part 1). 2002. Criteria for earthquake resistant design of structures. New Delhi: Bureau of
Indian Standards.
IS 456. 2000. Plain and reinforced concrete-code of practice, Bureau of Indian Standards, New Delhi.
New Delhi: Bureau of Indian Standards.
IS 875 (part 1). 1987. Code of practice for design loads (other than earthquake) for buildings and
structures — dead loads. New Delhi: Bureau of Indian Standards.
IS 875 (part 2). 1987. Code of practice for design loads (other than earthquake) for buildings and
structures — live loads. New Delhi: Bureau of Indian Standards.
Jalayer, F., and C. A. Cornell. 2003. A technical framework for probability-based demand and
capacity factor (DCFD) seismic formats. RMS.
Jayaram, N., L. Ting, and J. W. Baker. 2011. A computationally efficient ground-motion selection
algorithm for matching a target response spectrum mean and variance. Earthquake Spectra 27
(3): 797–815. doi:10.1193/1.3608002.
Jehel, P., P. Léger, and A. Ibrahimbegovic. 2014. Initial versus tangent stiffness-based Rayleigh
damping in inelastic time history seismic analyses. Earthquake Engineering & Structural
Dynamics 43 (3): 467–84. doi: 10.1002/eqe.v43.3.
Jeong, S.-H., and A. S. Elnashai. 2004. Analytical assessment of an irregular RD full scale 3D test
structure. Urbana: University of Illinois at Urbana-Champaign.
Kircher, C., G. Deierlein, J. Hooper, H. Krawinkler, S. Mahin, B. Shing, and J. Wallace. 2010.
Evaluation of the FEMA P-695 methodology for quantification of building seismic performance
factors. No. NIST GCR 10-917-8. Gaithersburg, MD: National Institute of Standards and
Technology.
Krawinkler, H., and A. A. Nassar. 1992. Seismic design based on ductility and cumulative damage
demands and capacities. In Nonlinear seismic analysis and design of reinforced concrete buildings,
ed. P. Fajfar, and H. Krawinkler, 23–39. Bled, Slovenia, Yugoslavia.
Liel, A. B., C. B. Haselton, and G. G. Deierlein. 2011. Seismic collapse safety of reinforced concrete
buildings. II: Comparative assessment of nonductile and ductile moment frames. Journal of
Structural Engineering 137 (4): 492–502. doi: 10.1061/(ASCE)ST.1943-541X.0000275.
26 P. S. BADAL AND R. SINHA

Lowes, L. N., N. Mitra, and A. Altoontash. 2003. A beam-column joint model for simulating the
earthquake response of reinforced concrete frames. Berkeley: University of California.
Luco, N., B. R. Ellingwood, R. O. Hamburger, J. D. Hooper, J. K. Kimball, and C. A. Kircher. 2007.
Risk-targeted versus current seismic design maps for the conterminous United States. In
Structural Engineers Association of California 76th Annual convention, Squaw Creek, California.
McKenna, F., G. Fenves, and M. Scott, others. 2000. Open system for earthquake engineering
simulation. Berkeley: University of California.
Meinheit, D. F., and J. O. Jirsa. 1981. Shear strength of R/C beam-column connections. Journal of
the Structural Division 107 (11): 2227–44.
Miranda, E., and V. V. Bertero. 1994. Evaluation of strength reduction factors for earthquake-
resistant design. Earthquake Spectra 10 (2): 357–79. doi: 10.1193/1.1585778.
Miyamoto, H. K., A. S. Gilani, A. Wada, and C. Ariyaratana. 2011. Identifying the collapse hazard of
steel special moment-frame buildings with viscous dampers using the FEMA P695 methodology.
Earthquake Spectra 27 (4): 1147–68. doi: 10.1193/1.3651357.
Mondal, A., S. Ghosh, and G. Reddy. 2013. Performance-based evaluation of the response reduction
factor for ductile RC frames. Engineering Structures 56: 1808–19. doi: 10.1016/j.
engstruct.2013.07.038.
Nobahar, E., M. Farahi, and M. Mofid. 2016. Quantification of seismic performance factors of the
buildings consisting of disposable knee bracing frames. Journal of Constructional Steel Research
124: 132–41. doi: 10.1016/j.jcsr.2016.05.007.
Panagiotakos, T. B., and M. N. Fardis. 2001. Deformations of reinforced concrete members at
yielding and ultimate. ACI Structural Journal 98 (2): 135–148.
Paulay, T., and M. Priestly. 1992. Seismic design of reinforced concrete and masonry buildings. New
York: Wiley.
PEER/ATC 72-1. 2010. Modeling and acceptance criteria for seismic design and analysis of tall
buildings. Vol. 113. Redwood City: Applied Technology Council, PEER.
Petrini, L., C. Maggi, M. N. Priestley, and G. M. Calvi. 2008. Experimental verification of viscous
damping modeling for inelastic time history analyzes. Journal of Earthquake Engineering 12 (S1):
125–45. doi: 10.1080/13632460801925822.
Porter, K. A. 2003. An overview of PEER’s performance-based earthquake engineering methodol-
ogy. Proceedings of Ninth International Conference on Applications of Statistics and Probability
in Civil Engineering, San Francisco, CA.
Ramirez, C., A. Liel, J. Mitrani-Reiser, C. Haselton, A. Spear, J. Steiner, G. Deierlein, and E.
Miranda. 2012. Expected earthquake damage and repair costs in reinforced concrete frame
buildings. Earthquake Engineering & Structural Dynamics 41 (11): 1455–75. doi: 10.1002/eqe.
v41.11.
Sato, A., and C.-M. Uang. 2010. Seismic performance factors for cold-formed steel special bolted
moment frames. Journal of Structural Engineering 136 (8): 961–67. doi: 10.1061/(ASCE)ST.1943-
541X.0000191.
Tesfamariam, S., M. Sánchez-Silva, and P. Rajeev. 2013. Effect of topology irregularities and
construction quality on life-cycle cost of reinforced concrete buildings. Journal of Earthquake
Engineering 17 (4): 590–610. doi: 10.1080/13632469.2012.762955.
Ulrich, T., C. Negulescu, and J. Douglas. 2014. Fragility curves for risk-targeted seismic design
maps. Bulletin of Earthquake Engineering 12 (4): 1479–91. doi: 10.1007/s10518-013-9572-y.
Vamvatsikos, D., and C. A. Cornell. 2002. Incremental dynamic analysis. Earthquake Engineering &
Structural Dynamics 31 (3): 491–514. doi: 10.1002/(ISSN)1096-9845.
Whittaker, A., G. Hart, and C. Rojahn. 1999. Seismic response modification factors. Journal of
Structural Engineering 125 (4): 438–44. doi: 10.1061/(ASCE)0733-9445(1999)125:4(438).
Whittaker, A., G. M. Atkinson, J. W. Baker, J. Bray, D. N. Grant, R. O. Hamburger, C. B. Haselton,
and P. G. Somerville. 2011. Selecting and scaling earthquake ground motions for performing
response-history analyses. No. NIST GCR 11-917-15. Gaithersburg, MD: National Institute of
Standards and Technology.
 JOURNAL OF EARTHQUAKE ENGINEERING 27

Appendix A. Validation of numerical model

The present study used a nonlinear two-dimensional numerical model with concentrated plasticity
to predict the seismic response of special reinforced concrete frame buildings. An analytical model
was developed using open-source simulation software OpenSees (McKenna, Fenves, and Scott
2000). Member model uses Ibarra-Medina-Krawinkler (IMK) hysteretic rules for beam-columns
(Ibarra, Medina, and Krawinkler 2005). The IMK model captures strength and stiffness degradation,
which are particularly crucial for simulating the response to long-duration time history records. Key
points on the backbone curve have been derived using semi-empirical expressions proposed in the
published literature (Haselton et al. 2008, 2016; Panagiotakos and Fardis 2001). Specifically, secant
stiffness to the 40% of yield moment, EIstf40 has been used for pre-yield stiffness. The beam-column
joints are modeled with a diagonal compression strut mechanism. This model matches well with the
experimentally observed cyclic response of the joints (Lowes, Mitra, and Altoontash 2003). The
joint shear panel stiffness is based on Meinheit and Jirsa (1981). The schematic sub-assemblage of
the analytical model is shown in Fig. 4. Geometric nonlinearity (P-Δ effect) has been captured using
a leaning column frame.
Special reinforced concrete frame buildings with capacity-based shear design considered in this
study has a predominant flexural failure and hence only the flexural hinges have been modeled.
Experimental results of a ductile moment-resisting frame subjected to seismic load from Filiatrault,
Lachapelle, and Lamontagne (1998) have been used to validate the nonlinear analytical model. More
details regarding material properties, section properties, and applied loading of the frame can be
found in Filiatrault, Lachapelle, and Lamontagne (1998).
Time history component N04W of the accelerogram recorded in Olympia, Washington, during
the April 13, 1949, Western Washington earthquake scaled up from the recorded peak horizontal
acceleration of 0.16 g to 0.21 g was applied for the first test (Intensity 1). Subsequently, the same
time history record but scaled to 0.42 g was applied to the damaged structure (Intensity 2). In the
time period range of interest, it was found that the mean difference between experimentally
observed and the simulated values are 8.5% for first intensity, and 2.8% for second intensity.

Figure A1. Top floor relative displacement time-history (Intensity 1 excitation).

28 P. S. BADAL AND R. SINHA

Table A1. Comparison of peak responses from the experimental study (Filiatrault, Lachapelle, and
Lamontagne 1998) and numerical model from the present study.
Experimental Numerical
Parameter Intensity 1 Intensity 2 Intensity 1 Intensity 2
First floor peak inter-story drift (%) 1.58 2.74 1.21 2.33
Second floor peak inter-story drift (%) 1.44 2.23 1.33 2.18
First Floor acceleration amplification factor 2.19 1.55 2.60 1.81
Second Floor acceleration amplification factor 4.30 2.45 4.63 3.01

Relative displacement and absolute acceleration time-history response of top floor for Intensity 1
obtained from the numerical model has been compared with the experimental observations in
Fig. A1. Similar results were also found for Intensity 2. The primary objective of the present study is
to accurately estimate the highly nonlinear peak responses. Table A1 presents a comparison of peak
responses from experimental data and the results of numerical model. It can be seen that a good
match between the experimental and numerical results is found, confirming that the nonlinear
analytical model used in the present study adequately predicts nonlinear responses.