See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/342107535

Determination of torsional irregularity in response spectrum analysis of

building structures

Article  in  Structural Engineering & Mechanics · June 2020

DOI: 10.12989/sem.2020.74.5.699

CITATIONS READS

2 726

3 authors, including:

Fatemeh Aliakbari Sadegh Garivani

Ferdowsi University Of Mashhad University of Bojnord
5 PUBLICATIONS   45 CITATIONS    15 PUBLICATIONS   156 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Fatemeh Aliakbari on 12 January 2022.

The user has requested enhancement of the downloaded file.

Structural Engineering and Mechanics, Vol. 74, No. 5 (2020) 699-709
DOI: https://doi.org/10.12989/sem.2020.74.5.699 699

Determination of torsional irregularity in response spectrum analysis of

building structures
Fatemeh Aliakbari1a, Sadegh Garivani 1 and Ali Shahmari2
1
Faculty of Engineering, University of Bojnord, Bojnord 94531-55111, Iran
2
Faculty of Civil and Environmental Engineering, Tarbiat Modares University, Tehran 14115-143, Iran

(Received March 24, 2019, Revised January 6, 2020, Accepted January 21, 2020)

Abstract. Torsional irregularity is one of the most probable types of horizontal irregularity and existence of this irregularity in
most of the structural loading codes is determined by calculating the ratio of the maximum to the average story drift. No specific
method has been previously recommended by the codes to calculate the mentioned ratio in the response spectrum analyses. In
the current investigation, nine steel building structures with different plan layouts and number of stories have been analyzed and
designed in order to evaluate the efficiency of three methods for calculating the ratio of the maximum to the average story drift
in the response spectrum analyses. It should be noted that one of these methods is the approach used by current version of
ETABS software andother ones are proposed in this paper. The obtained results using the proposed methods are compared with
the time history analysis results. The comparisons show that one of these methods underestimates the mentioned ratio in all
studied models, however, the other two methods have shown similar results. It is also found that the plan layouts and
irregularities can affect how these methods estimate the ratios compared to those obtained by the time history analysis.
Generally, it can be concluded that all of these methods can properly predict the ratio with acceptable errors.
Keywords: response spectrum analysis; Irregular structures; Horizontal irregularity; Torsional irregularity; Extreme
torsional irregularity

1. Introduction were also reported in other earthquakes that were due to

their irregularity and asymmetry (Elnashai et al. 2010,
Earthquakes are one of the inevitable natural phenomena
Wyllie et al. 1986). The behavior examination of buildings
and it is impossible to control all of its outcomes. However,
with reentrant corner irregularity has shown the unsuitable
structural design codes are used to predict and control the
performance of these structures against severe earthquakes
seismic behavior of structures. The existence of
(Penelis et al. 1988, Penelis and Kappos 1997). Vertical
irregularities in structures increases the probability of
irregularities result in story failures because of non-uniform
seismic structural damages. Considering aesthetic matters in
distribution of damage states along the height. However,
design of various ranges of buildings, from residential to
plan irregularities result in non-uniform damage states
theaters and museums, imposes irregularity and asymmetry
among the columns of a floor (Jeong and Elnashai 2006).
to the architectural design of buildings. The seismic
Seismic code provisions are constantly being updated to
behavior of irregular structures differs from that of regular
improve the seismic behavior of the designed structures.
structures and is more complicated. Thus, seismic loading
The updates are based on the gained experience from real
codes classify the buildings in two groups of regular and
earthquakes and the results of extensive researches carried
irregular buildings. The irregularity of structures is mainly
out to control and to develop the rules of the standards (e.g.
due to the asymmetric distribution of mass, strength and
recent research works done in the fields of calculating the
stiffness. The performance examination of the structures
fundamental period of structures (Asteris et al. 2015,
during the past earthquakes has shown that this feature can
Asteris et al. 2017, Asteris and Nikoo 2019, Harris and
be the main source of severe damages and the irregular
Michel 2019), response modification factor (Aliakbari and
structures can be prone to more damage compared to the
Shariatmadar 2019, Avanaki 2019, Mohsenian et al. 2019),
normal regular buildings (Jeong and Elnashai 2006). In the
etc.).
Mexico City earthquake in 1985, for example, it was
Numerous studies have also been dedicated to the
observed that the buildings with horizontal irregularity were
effects of the torsional irregularities on seismic behavior of
damaged more than the regular ones (Rosenbleuth and Meli
the structures in recent years. Some researchers have
1986). Similar patterns of damage and collapse of buildings
focused on the nonlinear behavior of irregular structures to
improve the push-over analysis methods of these structures.
Corresponding author, Assistant Professor Modal Push-over analysis method has been successfully
E-mail: [email protected] adopted to estimate and simulate the seismic demand of
a
M.Sc. Lecturer asymmetrical buildings with proper accuracy (Chopra and
b
Ph.D. Graduate Goel 2004). However, in irregular buildings, it has been
shown that results of push-over analysis method differ from
Copyright © 2020 Techno-Press, Ltd.
http://www.techno-press.com/journals/sem&subpage=7 ISSN: 1225-4568 (Print), 1598-6217 (Online)
700 Fatemeh Aliakbari, Sadegh Garivani and Ali Shahmari

those obtained by the nonlinear dynamic analysis method. (the ratio of the maximum displacement drift of a floor
But, the results of the modified push-over analysis method corner to the average displacement drift of the considered
presented by Fajfar et al. (2005) are consistent with the edge of the floor). These parametric analyses showed that
responses of nonlinear dynamic analysis (D'Ambrisi et al. the dispersion with regard to the ratio between the
2009). A new pushover method was offered to estimate the eccentricity, which is parallel to the excitation direction and
seismic responses of asymmetrical structures in the plan in the radius of gyration is reduced if the modified Q ratio is
which the load was distributed in the height based on the used instead of the code proposed parameter. Demir et al.
combination of modal torsional moment and shear force of (2010) investigated the effective torsional irregularity
the stories (Shakeri et al. 2014). A robust computer code factors in six types of multi storey structures with shear
was developed and implemented to perform the nonlinear wall-frame system and various story numbers, plan views
static analysis of the asymmetric reinforced concrete and shear wall locations. Gokdemir et al. (2013) studied the
structures (Mazza 2014). Several studies have also been torsional irregularity effects on the behavior of structures
conducted to establish and develop consistent analytical during earthquakes. They compared the base shear forces
methods in order to capture the deformational behavior of and torsion values in various irregular structures and also
irregular structures (Penelis and Kappos 2002, Dogangun evaluated some provisions of seismic codes about the
and Livaoglu 2006, Jinjie et al. 2008, Mahdi and Gharaie structures with torsional irregularities. The results showed
2011). that splitting the large parts of a building into separate
In addition, some researches have focused on the blocks or increasing the lateral stiffness of the weaker side
behavior of irregular structures and control/development of of the structures could decrease the undesirable torsional
existing provisions for this type of structures. Given both effects. A parametric study was performed on six groups of
the serviceability and the ultimate limit states, an optimized typical structures with varying shear wall positions, number
procedure have been developed for the design of torsionally of stories and axis numbers to investigate the effect of
unbalanced structures subjected to earthquake excitations torsional irregularities on the deformational behavior of
by Duan and Chandler (1997). The effects of torsion on multi-story buildings under the earthquake excitations
the moment and the shear values of vertical structural (Ozmen and Gulay 2014). The validity of code provisions
elements of irregular structures have been investigated in a was discussed and a new provisional definition for torsional
parametric study. The typical studied buildings of this irregularity coefficient based on floor rotations was
investigation were analyzed by both the methods of proposed. The accuracy of the empirical formulas of the
equivalent earthquake loading and dynamic analysis and the fundamental period proposed in seismic codes was
results were compared (Tezcan and Alhan 2001). Ozmen examined in tall steel concentrically braced frames (Young
and Gulay (2002) performed a parametric study to observe and Adeli 2016). Three types of irregularities including
the amplification of internal forces in torsionally irregular vertical irregularity, horizontal irregularity and a
multi-story buildings and concluded that more rigorous combination of vertical and horizontal irregularity were
measurement are required in the codes for this rather critical considered in the study. The results showed that regular
type of irregularity. An advanced analytical assessment structures have larger periods compared to irregular ones
methodology has been presented for irregular buildings and this difference was less in the structures with horizontal
(Jeong and Elnashai 2004). Bosco et al. (2004) described irregularities. As a result, a new formula was proposed to
the results of a study devoted to the definition of the limits calculate the fundamental period of irregular eccentrically
of application of an approximated design method of non- braced tall steel frame structures (Young and Adeli 2016).
regularly asymmetric systems. They have shown that some The effect of the vertical geometric irregularities on the
clear limits could be defined in seismic codes for the fundamental periods of masonry infilled RC structures has
applicability of simplified approaches on irregular been investigated and the results showed that the
structures. The differences between the torsional irregularity fundamental period of the irregular building frames was
provisions in codes of China, USA and Europe have been smaller than the period of the regular building frames with
studied and it was indicated that torsion effects have no the same parameters (Asteris et al. 2017). The effect of the
correlation with the criterion adopted by the considered ratio Ω which is structure uncoupled torsional frequency,
codes and some regulations of these codes are not 𝜔𝜃 , to uncoupled transition frequency, 𝜔𝑥 or 𝜔𝑦 ) on the
reasonable (Zheng et al. 2004). The behavior of some torsional behaviour of the structure has been studied and the
reinforced concrete structures with extreme torsional results showed that structures with Ω<1.0 are more sensitive
irregularity in the plan has been studied under 56 ground to the value of eccentricity between center of mass and
motion records using nonlinear time history analysis to center of rigidity (Mehana et al. 2019). Mohsenian et al.
evaluate the accuracy of elastic analysis in estimation of (2019) have evaluated the seismic behavior of tunnel-form
plastic rotations. It was concluded that the response buildings with the horizontal irregularity and developed
spectrum analysis could accurately estimate the plastic appropriate design methodologies. Fragility curves were
rotations for the buildings with a short fundamental period also derived for various levels of intensity and simple
(Kosmopoulos and Fardis 2007). Ozhendekci and Polat equations were introduced to estimate the uncoupled
(2008) introduced a new parameter, Q, which is the ratio of frequency ratios. The results demonstrated that the torsional
horizontal to torsional effective modal masses to define the behavior of this type of structures was flexible and their
torsional irregularity of the buildings. Parametric analyses seismic capacity was adequate. Other studies have also been
were used for comparing Q with the ratio proposed in codes conducted to investigate the response of irregular structures
 Determination of torsional irregularity in response spectrum analysis of building structures 701

(Lee et al. 2011, Bigdeli et al. 2014, Landi et al. 2014). amount of each story drift in response spectrum analysis. In
In order to reduce the seismic hazards of irregular proposed method, the final value of any response is
structures compared to regular structures, the seismic design determined in different modes. Since displacements
codes have attempted to classify the structures into regular obtained from the modal analysis are the square root of the
and irregular and then, impose more stringent provisions for sum of squares (SRSS) combinations of the response of
irregular structures. For example, the ASCE7-16 (ASCE individual modes, the algebraic summation of them is not
2016) has classified irregularity of the buildings into two mathematically correct and it is not possible to calculate the
classes of horizontal and vertical irregularities and presents ratio of the maximum to the average story drift directly
special provisions for both of them. Horizontal irregularity from the combined displacements obtained from the modal
includes torsional and extreme torsional irregularities, analysis. On the other hand, since the ratio of the maximum
reentrant corner, diaphragm discontinuity, nonparallel to the average story drift is always greater than 1.0, the
system, out-of-plane offset irregularity. Vertical irregularity, obtained SRSS value of this ratio results in an
however, includes stiffness–soft story, stiffness–extreme unrealistically large value which is inaccurate. For example,
soft story, weight (mass), vertical geometric, in-plane this value will always be larger than or equal to the square
discontinuity in vertical lateral force-resisting element, root of the number of combined modes in the SRSS
discontinuity in lateral strength–weak story and combination method. Therefore, it seems that the discussed
discontinuity in lateral strength–extreme weak story ratio should be calculated using a reliable method in the
irregularity. The main behavioral difference between response spectrum analysis and this paper attempts to give
irregular and regular structures in the plan is the an appropriate response to this need.
significance of effects of torsional mode in the dynamic
behavior of the irregular structures. Importance of the
torsional modes depends on the amount of eccentricity of 3. Torsional irregularity and the importance of
the mass and stiffness in the structure. One of the most determining the ratio of maximum to average story
possible irregularities in the structures is torsional or drift
extreme torsional irregularities. According to the ASCE7-16
(ASCE 2016), if the maximum relative displacement at one In ASCE7-16 (ASCE 2016), torsional irregularity is
end of the building is larger than 20% of the average classified into two classes with the following definitions.
relative displacement at both ends of the building in a 1a. Torsional Irregularity: Torsional irregularity is
structure, the structure has torsional irregularity andif the defined to exist where the maximum story drift, computed
difference is more than 40%, the building has extreme including accidental torsion with 𝐴𝑥 = 1.0, at one end of
torsional irregularity. In recent years, many studies have the structure transverse to an axis is more than 1.2 times the
been conducted in order to predict the behavior and also to average of the story drifts at the two ends of the structure.
analyze and to design the irregular structures more Torsional irregularity requirements in the reference sections
accurately. apply only to structures in which the diaphragms are rigid
As discussed, the type of torsional irregularity is derived or semi-rigid.
from the ratio of the maximum to the average amount of 1b. Extreme Torsional Irregularity: Extreme torsional
each story drift, based on the provisions of the seismic irregularity is defined to exist where the maximum story
loading codes. So far, no method is presented to determine drift, computed including accidental torsion with 𝐴𝑥 = 1.0,
this ratio in response spectrum analysis in the codes. This at one end of the structure transverse to an axis is more than
paper examines the accuracy of some proposed methods for 1.4 times the average of the story drifts at the two ends of
calculating the ratio of maximum to average story drift in the structure. Extreme torsional irregularity requirements in
response spectrum analysis. To compare the results of these the reference sections apply only to structures in which the
methods with the results of time history analysis, some diaphragms are rigid or semi-rigid.
different structures in the plan and height have modeled and It should be mentioned that the extreme torsional
finally, the more suitable method is introduced. irregularity was not defined in the older codes, but
according to the importance of the issue, the subject is
2. Research significance explained with more precision. As it is observed, the ratio of
maximum to average story drift determines the type of the
In equivalent static methods, the ratio of the maximum torsional irregularity in the structure. It is very important to
to the average amount of each story drift can be directly determine the type of irregularity that depends on the value
calculated using the static analyses results. But the use of of this ratio because in some cases, the provisions of the
these methods for the analysis of irregular structures has structural loading change depending on the type of
many limitations, which often leads to the necessity of irregularity of the structure. For instance, the value of this
using dynamic analysis methods, especially in the structure ratio will be decisive to scale the design values of combined
with torsional and extreme torsional irregularities. In the responses obtained by the response spectrum analysis and
structural design process, it is essential to perform all design also to select the allowed analysis procedure. Another case
controls based on the results of the same method. While, as that shows the importance of determining the correct value
mentioned above, there is no robust and well-known of this ratio is to calculate the torsional amplification factor,
method to calculate the ratio of the maximum to the average 𝐴𝑥 . This factor can be calculated using Eq. (1) as follows.
702 Fatemeh Aliakbari, Sadegh Garivani and Ali Shahmari

software use this method to compute the ratio (CSI

Knowledge Base 2013).

𝐷𝑚𝑎𝑥 2
∑𝑛𝑖 ( ) (2)
√ 𝐷𝑎𝑣𝑒 𝑖
𝑅𝑎𝑡𝑖𝑜 =
𝑛

where 𝑖 and 𝑛 are the mode number and the total number
of modes considered in the spectrum dynamic, respectively.
Method 2: For each floor of the structure, the combined
drift values are obtained using the SRSS method at the end
joints of the structure (the joints A and B in Fig. 1) and then,
the maximum to average drift ratio is calculated (Eq. (3)).
Fig. 1 Torsional Amplification Factor, 𝐴𝑥 (ASCE 2016)
𝐷𝑚𝑎𝑥
𝑅𝑎𝑡𝑖𝑜 = (3)
𝐷𝑎𝑣𝑒

𝛿𝑚𝑎𝑥 2 where 𝐷𝑚𝑎𝑥 and 𝐷𝑎𝑣𝑒 are the maximum and average
𝐴𝑥 = ( ) (1) drifts, respectively, which are calculated directly using the
1.2𝛿𝑎𝑣𝑒
SRSS method. It should be noted that since in this method,
where 𝛿𝑚𝑎𝑥 and 𝛿𝑎𝑣𝑒 are the maximum and average of the drift values are used to compute the average drift
the displacement at level x computed assuming 𝐴𝑥 =1.0, obtained from the combination of the modal responses, the
respectively (Fig. 1). It should be mentioned that since the method is not mathematically acceptable. However, as it is
average and maximum displacement values are both popular among the designers due to its simplicity, the
divided into the story height to obtain the amount of story results of the method are investigated in this paper.
drift, the maximum to average ratio is equal for both Method 3: For each floor of the structure, the maximum
displacement and drift. and average drift values at the end joints of the structure are
calculated for each mode and then the SRSS combination of
the maximum and average drift is obtained separately (the
4. The proposed methods to evaluate/check the joints A and B in Fig. 1). Then, these values are divided by
torsional irregularity each other as presented Eq. (4).

In this paper, three methods are examined to calculate √∑𝑛𝑖(𝐷𝑚𝑎𝑥 )𝑖 2

the ratio of the maximum to the average story drift based on 𝑅𝑎𝑡𝑖𝑜 = (4)
the displacement responses obtained by the response √∑𝑛𝑖(𝐷𝑎𝑣𝑒 )𝑖 2
spectrum analysis as explained below. In all of these
methods, the SRSS method is used to combine the
responses of individual modes. It should be noted that it is
not the only criterion of combination that can be adopted in
modal analyses according to the literature and codes. For 5. Description of the examined structures and the
example, based on the ASCE 7-16, the value for each analysis method
parameter of interest calculated for the various modes
should be combined using the square root of the sum of the To examine the accuracy of the proposed methods to
squares (SRSS) method, the complete quadratic compute the maximum to average story drift in response
combination (CQC) method, the complete quadratic spectrum analysis, nine different three-dimensional steel
combination method (CQC-4), or an approved equivalent structures are modeled, analyzed and designed using
approach. According to the specifications of this code, ETABS (CSI 2016) software. These models include 4, 8 and
displacements/drifts are one of the responses that can be 12-story structures with three types of plans shown in Fig.
combined with the SRSS method. However, in this paper, 2. In this figure, CM shows the center of mass location in
considering the simplicity of the SRSS method compared to the plan. In Fig. 2(a), the dashed lines represent the X-
other ones, it is used to combine the displacement responses braced bays in the plan. It should be noted that the studied
of individual modes. structures do not have any vertical irregularities and all
Method 1: For each floor of the structure, the ratio of stories have an equal height of 3.2 m in all models. The
the maximum to the average story drift is obtained in each imposed Dead and Live loads are summarized in Table 1.
mode and then these values are combined using the SRSS The floor diaphragms were assumed to be rigid and the
method. Finally, the combined value is divided by the structural steel materials were assumed to be ST37 with the
square root of the number of modes. In other words, the yield stress and Young modulus equal to 240 MPa and 204
square root of the mean of squares of the maximum to GPa, respectively. In addition, it should be noted that the
average drift is computed in this method (Eq. (2)). It should analyses were performed in the Y-direction of Plan Types 1
be mentioned that the recent versions of ETABS (CSI 2016) and 2 and in the X-direction of Plan Type 3. However, in
 Determination of torsional irregularity in response spectrum analysis of building structures 703

Plan Type 2, considering the plan configuration, there is no 4@6m

difference in the direction of analysis.
The gravity and seismic loads were determined and
imposed in accordance with the ASCE 7-16 specifications
and the structural members were designed according to the
AISC360-10 provisions as well. It should be noted that the
response spectrum analysis was adopted as per the ASCE7-
16 to check the seismic performance of the structures. The
models designation along with other properties are given in

4@6m
CM
Table 2. As presented in this table, the lateral-force-resisting
systems considered for Plan Type 1 and Plan Types 2 and 3
are ordinary concentrically braced frame and intermediate
moment frame, respectively. It should be noted that the
lateral-force-resisting systems and their component
distributions in the plan, as well as the shape of plans, have
been selected and arranged so that the resulting structures Y
have the torsional and extreme torsional irregularities. For
example, in Plan Type 1, with the aim of having a structure x
with extreme-torsional irregularity, the symmetrical
distribution of the mass in the plan was considered with an Plan Type 1
asymmetric distribution of stiffness. Also, for the other plan 4@6m
types, the moment frame system is intended to have
structures with torsional irregularity. In this table, 𝑇1 , 𝐶𝑠 , R
and M are the fundamental period, seismic response
coefficient in the equivalent static method (used for scaling
design values of combined responses), the seismic response
modification factor (ASCE 2016) and the seismic mass of
the structure, respectively. The adopted spectrum used for
the seismic design of the frames is derived based on ASCE CM
4@6m

7-16 (ASCE 2016). It is assumed that this building will be

constructed in California for which 𝑆𝑠 = 1.0 𝑔 and 𝑆1 =
0.6 𝑔 . 𝑆𝑠 and 𝑆1 are spectral response acceleration
parameters at short periods and the period of 1 sec,
respectively. For the soil condition, it is assumed that site
class C is appropriate as shown in Fig. 8. The cross-sections
of the beams, columns and braces are I-shaped, built-up Y
Box and double-UNP sections, respectively. For example,
x
the designed structural members of Model 8-2 are listed in
Table 3. Plan Type 2
In order to examine the accuracy of the proposed
methods to determine the ratio of the maximum to average 4@6m
story drift explained above, the response of the structures is
obtained under seven artificial records. These records are
generated by SeismoArtif software in a way that the
average of their response spectrum is above the design
spectrum in the range of 0.2T-1.5T for all studied
structures, where T is the fundamental period of the
structure. The main records features are given in Table 4 CM
4@6m

and the comparison of their response spectrums and design

spectrum are given in Fig. 3.

Table 1 The assumed gravity loads in the structural models

Peripheral walls
Live load
Dead load Partitions load load Y
(kg/m2)
(kg/m2) (kg/m2) (kg/m)
floors roof floors roof x
500 200 150 100 580 250 Plan Type 3
Fig. 2 Three types of plan in studied structures
704 Fatemeh Aliakbari, Sadegh Garivani and Ali Shahmari

Table 2 The properties of studied structures

Model ID Plan Type No. of stories Seismic Force-Resisting System 𝑅 𝑇1 (𝑠) 𝑀 (𝑘𝑔) 𝐶𝑠
Model 4-1 Type 1 4 OCBF* 3.25 0.66 1810694 0.25
Model 4-2 Type 2 4 IMF** 4.50 0.79 1420857 0.18
Model 4-3 Type 3 4 IMF 4.50 0.93 1465539 0.18
Model 8-1 Type 1 8 OCBF 3.25 0.71 3722970 0.25
Model 8-2 Type 2 8 IMF 4.50 1.27 2883500 0.13
Model 8-3 Type 3 8 IMF 4.50 1.24 3014341 0.13
Model 12-1 Type 1 12 OCBF 3.25 1.13 5685558 0.23
Model 12-2 Type 2 12 IMF 4.50 1.70 4394245 0.09
Model 12-3 Type 3 12 IMF 4.50 1.73 4602335 0.09
*
Ordinary concentrically braced frame
**
Intermediate moment frame

Table 3 Sectional properties of Models 8-1, 8-2 and 8-3

Beam Column
Model Story Not Adjacent to Brace
Web plate Flange plate Adjacent to Brace
Brace
1 PL320×6 PL220×10 BOX620x35 BOX280X15 2UPN320
2 PL320×6 PL220×10 BOX620x35 BOX280X15 2UPN300
3 PL320×6 PL220×10 BOX620x35 BOX280X15 2UPN300
4 PL320×6 PL220×10 BOX560X30 BOX280X15 2UPN280
8-1
5 PL320×6 PL220×10 BOX560X30 BOX270X10 2UPN260
6 PL320×6 PL220×10 BOX450X25 BOX270X10 2UPN240
7 PL320×6 PL220×10 BOX450X25 BOX270X10 2UPN200
8 PL320×6 PL220×10 BOX450X25 BOX270X10 2UNP180
1 PL600×8 PL220×12 Box450×450×25 ----
2 PL600×8 PL220×12 Box450×450×25 ----
3 PL600×8 PL220×12 Box450×450×25 ----
4 PL600×8 PL220×12 Box340×340×20 ----
8-2
5 PL600×8 PL220×12 Box340×340×20 ----
6 PL600×8 PL220×12 Box340×340×20 ----
7 PL400×6 PL220×10 Box280×280×15 ----
8 PL320×6 PL220×10 Box280×280×15 ----
1 PL600×8 PL220×12 Box450×450×25 ----
2 PL600×8 PL220×12 Box450×450×25 ----
3 PL600×8 PL220×12 Box450×450×25 ----
4 PL600×8 PL220×12 Box340×340×20 ----
8-3
5 PL600×8 PL220×12 Box340×340×20 ----
6 PL600×8 PL220×12 Box340×340×20 ----
7 PL400×6 PL220×10 Box280×280×15 ----
8 PL320×6 PL220×10 Box280×280×15 ----

6. Precision control of the proposed methods maximum to average story drift for all the proposed
methods and studied models are shown in Fig. 4.
To evaluate the accuracy of the proposed methods, the Method 2 and 3 would give similar results, however,
ratios calculated using these methods are compared with one cannot mathematically and dynamically interpret the
those ones obtained by the linear time history dynamic former approach results despite its simplicity. Therefore,
analysis which conducted using the Hilber-Hughes-Taylor Method 2 can be used with reasonable accuracy for rough
method to linear direct integration. The ratios of the estimation purposes. In addition, the results show that
 Determination of torsional irregularity in response spectrum analysis of building structures 705

9
1.2

ASCE7 8 Method 2
1
Scaled Averag of Spectrums 7 Method 3

6 Method 1
0.8

Story Number
5 Time History
Sa(g)

0.6 4

3
0.4
2

0.2 1

0
0 1.45 1.47 1.49 1.51 1.53 1.55 1.57
0 0.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 Maximum Drift to Average Drift Ratio
T(s)
(d) Model 8-1
Fig. 3 Design spectrum and response spectra of the 9

selected artificial records 8 Method 2

7 Method 3
Table 4 Artificial records data list 6 Method 1

Story Number
Record PGA (g) 5-95% Duration (sec) 5 Time History

Artificial-01 0.61 6.79

4

3
Artificial-02 0.52 6.37
2
Artificial-03 0.48 7.79 1
Artificial-04 0.54 7.63 0
Artificial-05 0.59 6.66 1.05 1.1 1.15 1.2 1.25 1.3
Maximum Drift to Avergae Drift Ratio
Artificial-06 0.50 7.99
(e) Model 8-2
Artificial-07 0.52 7.59 9

8 Method 2
5
Method 2 7 Method 3

4 Method 3 6 Method 1
Story Number

Method 1 5 Time History

3
Story Number

Time History 4

3
2
2

1
1
0
1.13 1.14 1.15 1.16 1.17 1.18 1.19 1.2 1.21 1.22 1.23
0 Maximum Drift to Average Drift Ratio
1.46 1.48 1.5 1.52 1.54 1.56
Maximum Drift to Average Drift Ratio (f) Model 8-3
(a) Model 4-1 13
12
5 Method 2
11
Method 2 10 Method 3
4 Method 3 9
Method 1
Story Number

8
Method 1
7 Time History
Story Number

3 Time History
6
5
2 4
3
2
1 1
0
1.35 1.4 1.45 1.5 1.55 1.6
0
1.08 1.09 1.1 1.11 1.12 1.13 1.14 1.15 Maximum Drift to Average Drift Ratio
Maximum Drift to Average Drift Ratio (g) Model 12-1
(b) Model 4-2 13
12
5 Method 2
11
Method 2 Method 3
10
4 Method 3 9 Method 1
8
Story Number

Method 1
Story Number

7 Time History
3
Time History
6
5
2
4
3
1 2
1
0
0
1.06 1.08 1.1 1.12 1.14 1.16 1.18
0.95 1 1.05 1.1 1.15 1.2

Maximum Drift to Average Drift Ratio Maximum Drift to Average Drift Ratio

(c) Model 4-3 (h) Model 12-2

706 Fatemeh Aliakbari, Sadegh Garivani and Ali Shahmari

13 2.50
12
Method 2
11 2.13
10 Method 3
9
2.00
Story Number

Method 1
8
7 Time History 1.53 1.53
6 1.50

Error (%)
5
4
3
2
1.00
1
0
1.17 1.19 1.21 1.23 1.25 1.27 1.29 1.31 0.50
Maximum Drift to Average Drift Ratio

(i) Model 12-3

Fig. 4 Ratios of maximum to average story drift in studied 0.00

models Method 1 Method 2 Method 3

(a) Model 4-1
3.50
Method 1 predicts smaller values for the ratio of the
2.96
maximum drift to the average drift ratios compared to 3.00
Method 2 and 3 except for Model 12-2 as well as some 2.52 2.60
2.50
lower stories in model 4-2 and model 4-3. Therefore, it
seems that if the methods 2 or 3 are used, the values of the 2.00
mentioned ratio are computed more conservatively and Error (%)
these methods offer stricter results. 1.50
On the other hand, the comparison of the ratios obtained
from proposed methods with the time history analysis 1.00

results shows that Method 1 underestimates the ratios in all 0.50

studied models. Therefore, it can be concluded that this
method is non-conservative for determining the ratio of 0.00
maximum to average story drift. In methods 2 and 3, Method 1 Method 2 Method 3
however, the results of the models with Plan Type 1 show
(b) Model 4-2
that the ratios are greater than those determined from time 6.00
history analysis. However, in all models with plan types 2 5.45
5.18 5.21
and 3, an opposite pattern is observed except Model 8-2. 5.00
Thus, it seems that the source of torsional irregularity
affects the results of the methods 2 and 3 and making a 4.00
definitive judgment about conservancy of these methods
needs more analysis.
Error (%)

3.00
For a better and more consistent comparison, the
average of absolute values of errors in each method are 2.00
compared to the time history responses and are shown in
Fig. 5 for all studied models. In addition, the values of 1.00
errors are computed separately in terms of the type of plan
and number of stories in Figs. 6-7, respectively. As seen in 0.00
these figures, methods 2 and 3 have lower errors compared Method 1 Method 2 Method 3
to method 1 in the models with the plans types 1 and 3.
However, method 1 has a lower error in the models with (c) Model 4-3
3.00
plan 2. This shows that the type of plan can directly affect
the error of the studied methods. In addition, it can be seen
2.50 2.36
that increasing the height of the structure increases the error
of the proposed methods. If the average of all frames is set
2.00
to the benchmark presented in Figs. 6-7, it can be concluded
that the absolute values of errors do not differ significantly
Error (%)

1.45 1.44
1.50
in the three proposed methods and hence it seems that
designers can simply use each of these methods although
1.00
the method 2 seems to be simpler and more straightforward
than other ones.
0.50

7. Conclusions 0.00
Method 1 Method 2 Method 3
In this paper, three different methods are introduced and (d) Model 8-1
examined to determine the ratio of the maximum to average
 Determination of torsional irregularity in response spectrum analysis of building structures 707

7.00 7.00
6.30
6.00
5.86 5.82 6.00

5.00 5.00

3.90 3.91
4.00 3.62 4.00

Error (%)
Error (%)

3.00 3.00

2.00 2.00

1.00
1.00

0.00
0.00
Method 1 Method 2 Method 3
Method 1 Method 2 Method 3
(e) Model 8-2 (i) Model 12-3
6.00 Fig. 5 Average of absolute values of the errors of
proposed method in all studied models
4.99
5.00

4.00
3.96 3.96 story drift in order to identify the torsional irregularity of
the structures in the response spectrum analyses and their
precision was examined. In this regard, numerical
Error (%)

3.00
simulations are performed and the response of the studied
2.00
structures was determined based on the time history and
response spectrum analyses. The results were obtained
1.00 using the proposed methods and compared with the
responses of the time history analyses. To sum up, the
0.00 following results were obtained:
Method 1 Method 2 Method 3 • Although Method 2 is not theoretically correct but is
(f) Model 8-3 the simplest approach and it can give similar and relatively
5.00 conservative results compared to Method 3.
4.50
4.52 • Compared to Method 1, the other methods can
estimate the ratio of the maximum to average story drift
4.00 3.77
3.62 larger and present stricter results.
3.50
• Method 1 underestimates the ratio of maximum to
3.00 average story drift and gives less conservative results.
Error (%)

2.50 • If the total average of absolute errors in all studied

2.00 models is considered as the acceptable criterion for design
1.50
purposes, the error of all proposed methods are close, so
designers can choose one of these methods depending on
1.00
their preferences.
0.50
• As a general suggestion, with consideration of all
0.00 aspects, Method 3 is suggested for calculating the ratio of
Method 1 Method 2 Method 3 maximum to average story drift.
(g) Model 12-1
14.00 13.28 References
12.73
12.00 AISC (2010), Specification for Structural Steel Buildings,
ANSI/AISC 360-10, Chicago.
10.00 ASCE (2016), Minimum design loads for buildings and other
structures, ASCE/SEI 7., ASCE, Reston, VA, USA.
Error (%)

8.00
Aliakbari, F. and H. Shariatmadar (2019), “Seismic response
6.00 modification factor for steel slit panel-frames”, Eng. Struct.,
4.85
181, 427-436. https://doi.org/10.1016/j.engstruct.2018.12.027.
4.00 Avanaki, M. J. (2019), “Response modification factors for seismic
design of steel Fiber Reinforced Concrete segmental tunnels”,
2.00 Constr Build Mater., 211, 1042-1049.
https://doi.org/10.1016/j.conbuildmat.2019.03.275.
0.00
ASCE (2016), Minimum design loads for buildings and other
Method 1 Method 2 Method 3 structures, ASCE/SEI 7., ASCE, Reston, VA, USA.
(h) Model 12-2 Asteris, P.G., Repapis, C.C., Tsaris, A.K., Di Trapani, F. and
Cavaleri, L. (2015), “Parameters affecting the fundamental
708 Fatemeh Aliakbari, Sadegh Garivani and Ali Shahmari

8.0
4ST 6.9
7.0 6.8
8ST
6.0 12ST
5.2
5.0 AVG 4.6
Error (%) 4.6
4.1
4.0 3.7 3.8 3.7
3.4 3.2
3.1
3.0

2.0

1.0

0.0
Method 1 Method 2 Method 3
Fig. 6. Average errors of proposed method in term of the number of stories

8.0
7.4
Plan-1 7.1
7.0
Plan-2
6.0 Plan-3 5.6
AVG
5.0 4.6 4.6
4.4 4.4
Error (%)

4.1
4.0 3.7
3.0
3.0
2.3 2.2
2.0

1.0

0.0
Method 1 Method 2 Method 3
Fig. 7. Average errors of proposed method in terms of the type of the plan

period of infilled RC frame structures”, Earth. Struct., 9(5), 999- Computers and Structures (CSI), (2016), “ETABS 2016 linear and
1028. http://dx.doi.org/10.12989/eas.2015.9.5.999. nonlinear static and dynamic analysis and design of three-
Asteris, P. G., Repapis, C. C., Foskolos, F., Fotos, A. andTsaris, A. dimensional structures”, Berkeley, CA.
K. (2017), “Fundamental period of infilled RC frame structures CSI Knowledge Base (Computers and Structures, Inc.), ETABS (2013),
with vertical irregularity”, Struct. Eng. Mech., 61(5), 663-674. Torsional Irregularities check for Response Spectrum Analysis (RSA),
https://doi.org/10.12989/sem.2017.61.5.663. https://wiki.csiamerica.com/pages/viewpage.action?pageId=4554971.
Asteris, P.G., Nikoo, M. (2019), “Artificial bee colony-based D'Ambrisi, A., De Stefano, M. andTanganelli, M. (2009), “Use of
neural network for the prediction of the fundamental period of pushover analysis for predicting seismic response of irregular
infilled frame structures”, Neural Comput. Appl., 31(9), 4837- buildings: A case study”, J. Earthquake Eng., 13(8), 1089-1100.
4847. https://doi.org/10.1007/s00521-018-03965-1. https://doi.org/10.1080/13632460902898308.
Bigdeli, Y., Kim, D. andChang, S. (2014), “Vibration control of Demir A, Demir D. D, Erdem R.T. and Bagci, M (2010),
3D irregular buildings by using developed neuro-controller “Torsional irregularity effects of local site classes in multiple
strategy”, Struct. Eng. Mech., 49(6), 687-703. storey structures”, Int J. Res. Rev. Appl. Sci., 258–262.
http://dx.doi.org/10.12989/sem.2014.49.6.687. Dogangun A. and Livaoglu, R. (2006), “Comparison of seismic
Bosco M, Marino E, Rossi P. P (2004), “Limits of application of analysis methods for multistory buildings”, First European
simplified design procedures to non-regular asymmetric conference on earthquake engineering and seismology, Paper
buildings”, 13th World Conference on Earthquake Engineering, No. 1314.4, September, Geneva, Switzerland.
Paper No. 886, Vancouver, Canada, August. Duan X. N, Chandler A. M (1997), “An optimized procedure for
Chopra, A. K. andGoel, R. K. (2002), “A modal pushover analysis seismic design of torsionally unbalanced structures”,
procedure for estimating seismic demands for buildings”, Earthquake Eng Struct Dynam., 26(7), 737–757.
Earthquake Eng. Struct. Dyn., 31(3), 561-582. https://doi.org/10.1002/(SICI)1096-9845(199707)26:7%3C737::AID-
https://doi.org/10.1002/eqe.144 EQE673%3E3.0.CO;2-S.
Chopra, A. K. andGoel, R. K. (2004), “A modal pushover analysis Elnashai, Amr S., Gencturk, B., Kwon, O., Al-Qadi, I.L., Hashash,
procedure to estimate seismic demands for unsymmetric‐plan Y., Roesler, J.R., Kim, S., Jeong, S., Dukes, J. and Valdivia, A.
buildings”, Earthquake Eng. Struct. Dyn., 33(8), 903-927. (2010), “The Maule (Chile) earthquake of February 27, 2010:
https://doi.org/10.1002/eqe.380 Consequence assessment and case studies”, MAE Center
 Determination of torsional irregularity in response spectrum analysis of building structures 709

Report No. 10-04, Department of Civil and Environmental Penelis, G.G. and Kappos, A. (1997), Earthquake-resistant
Engineering, University of Illinois at Urbana-Champaign, USA. Concrete Structures, E & FN SPON (Chapman & Hall),
Fajfar, P., Marušić, D. and Peruš, I. (2005), “Torsional effects in London, United Kingdom.
the pushover-based seismic analysis of buildings”, J. Rosenbleuth, E. and Meli, R. (1986), “The 1985 earthquake:
Earthquake Eng., 9(06), 831-854. Causes and effects in Mexico City”, Concert International, 8(5).
Gokdemir, H., Ozbasaran, H., Dogan, M., Unluoglu, E. Penelis, G. and Kappos A. J. (2002), “3D pushover analysis: the
andAlbayrak, U. (2013), “Effects of torsional irregularity to issue of torsion”, 12th European conference on earthquake
structures during earthquakes”, Eng. Fail. Anal., 35, 713-717. engineering, Paper No. 015, London, United Kingdom.
https://doi.org/10.1016/j.engfailanal.2013.06.028. Tezcan, S.S. and Alhan, C. (2001), “Parametric analysis of
Harris, J.L. and Michel, J.L. (2019), “Approximate Fundamental irregular structures under seismic loading according to the new
Period for Seismic Design of Steel Buildings Assigned to High Turkish Earthquake Code”, Eng. Struct., 23, 600-609.
Risk Categories”, Practice Periodical on Structural Design and https://doi.org/10.1016/S0141-0296(00)00084-5.
Construction, 24(4). https://doi.org/10.1061/(ASCE)SC.1943- Turkish Earthquake Code (2007), “TEC2007, Specification for
5576.0000444. buildings to be built in seismic zones”, Ministry of Public Works
Jeong, S.H. and Elnashai A.S. (2004), “Analytical and and Settlement, government of Republic of Turkey, Turkey.
experimental seismic assessment of irregular RC buildings”, SeismoArtif(ver. 2.1.0), A Program Process Strong-Motion Data,
13th World Conference on Earthquake Engineering, Paper No. http://seismisoft.com/seismoartif.
113, Vancouver, Canada, August. Shakeri, K., Tarbali, K. and Mohebbi, M. (2014), “Pushover
Jeong, S.H. and Elnashai, A.S. (2006), “New three-dimensional analysis of asymmetric-plan buildings based on distribution of
damage index for RC buildings with planar irregularities”, J. the combined modal story shear and torsional moment”,
Struct. Eng., ASCE, 132(9), 1482-1490. Earthquake Eng. Eng. Vib., 13(4), 707-716.
https://doi.org/10.1061/(ASCE)0733-9445(2006)132:9(1482). https://doi.org/10.1007/s11803-014-0274-5.
Jinjie, M., Qingxuan, S. and Qi, Z. (2008), “Method of Wyllie, L.A., Abrahamson N., Bolt, B., Castro, G., Durkin, M.E.,
performance based seismic evaluation for irregular plane Escalante, L., Gates, H.J., Luft, R., McCormick, D., Olson,
reinforced concrete frame structures”, 14th World Conference R.S., Smith, P.D. and Vallenas, J. (1986), “The Chile
on Earthquake Engineering, Beijing, China, October. Earthquake of March 3, 1985”, Earthquake Spectra, 2(2), 293-
Kosmopoulos, A. andFardis, M. (2007), “Estimation of inelastic 371.
seismic deformations in asymmetric multistorey RC buildings”, Young, K. and Adeli, H. (2016), “Fundamental period of irregular
Earthquake Eng. Struct. Dyn., 36(9), 1209-1234. eccentrically braced tall steel frame structures”, J. Constr. Steel
https://doi.org/10.1002/eqe.678 Res., 120, 199-205. https://doi.org/10.1016/j.jcsr.2016.01.001.
Landi, L., Pollioa, B. and Diotallevi, P.P. (2014), “Effectiveness of Zheng N, Yang Z, Shi C, Chang Z (2004). “Analysis of criterion
different standard and advanced pushover procedures for regular for torsional irregularity of seismic structures”, 13th World
and irregular RC frames”, Struct. Eng. Mech., 51(3), 433-446. Conference on Earthquake Engineering, Paper No. 1465,
https://doi.org/10.12989/sem.2014.51.3.433. Vancouver, Canada, August.
Lee, H.S., Jung, D.W., Lee, K.B., Kim, H.C. and Lee, K. (2011),
“Shake-table responses of a low-rise RC building model having
irregularities at first story”, Struct. Eng. Mech., 40(4), 517-539. CC
https://doi.org/10.12989/sem.2011.40.4.517
Mahdi T. and Gharaie V.S. (2011), “Plan irregular RC frames:
comparison of pushover with nonlinear dynamic analysis”,
Asian J Civil Eng Build Housing, 12(6), 679–690.
Mazza, F. (2014), “Modelling and nonlinear static analysis of
reinforced concrete framed buildings irregular in plan”, Eng. Struct.
J., 80, 98-108. https://doi.org/10.1016/j.engstruct.2014.08.026
Mehana, M.S., Mohamed, O. and Isam, F. (2019), “Torsional
Behaviour of Irregular Buildings with Single Eccentricity”, IOP
Conference Series: Materials Science and Engineering, 603(5),
052028.
Mohsenian, V., Nikkhoo, A. and Hajirasouliha, I. (2019),
“Estimation of seismic response parameters and capacity of
irregular tunnel-form buildings”, Bull. Earthq. Eng., 17(9),
5217-5239. https://doi.org/10.1007/s10518-019-00679-0.
Ozmen, G. and Gulay, F.G. (2002), “An investigation of
torsionally irregular multi-story buildings under earthquake
loading”, Struct. Eng. Mech., 14(2), 237-243.
https://doi.org/10.12989/sem.2002.14.2.237.
Ozhendekci, N. and Polat, Z. (2008), “Torsional irregularity of
buildings”, 14th World Conference on Earthquake Engineering,
Beijing, China, October.
Ozmen, G., Girgin, K., Durgun, Y. (2014), “Torsional irregularity
in multi-story structures”, Int. J. Adv. Struct. Eng., 6(4), 121-
131. https://doi.org/10.1007/s40091-014-0070-5.
Penelis, G. G., Sarigiannis, D., Stavrakakis, E. andStylianidis, K.
C. (1988), “A statistical evaluation of damage to buildings in the
Thessaloniki, Greece, earthquake of June, 20, 1978”, 9th World
Conf. on Earthquake Engineering, 7, 187-192. Kyoto-Tokyo,
Japan.

View publication stats