Send Orders of Reprints at reprints@benthamscience.

org

200 The Open Civil Engineering Journal, 2012, 6, (Suppl 1-M8) 200-214

Open Access

The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry
Structures

A.H. Akhaveissy*

Department of Civil Engineering, Faculty Engineering, Razi University, P.O. Box: 6714967346, Kermanshah, Iran

Abstract: A nonlinear finite element method with eight-noded isoparametric quadrilateral elements is used to predict the
behavior of unreinforced masonry structures. The disturbed state concept (DSC) with modified hierarchical single yield
surface (HISS) plasticity which is called DSC/HISS-CT is used to characterize the constitutive behavior of masonry in
both compression and tension. The model uses two HISS yield surfaces for compressive and tensile behavior. The DSC
model allows for the characterization of non-associative behavior through the use of disturbance. It computes microcrack-
ing during deformation, which eventually leads to fracture and failure. the critical disturbance, Dc, identifies fracture and
failure. In the DSC model the DSC model was validated at two levels: (1) specimen and (2) practical boundary value
problem. At the specimen level, predictions are obtained by integrating the incremental constitutive relations. The pro-
posed constitutive model is verified by comparing numerical predictions with results obtained from test data; the compari-
sons are found to be highly satisfactory. A new explicit formula is also presented to estimate the strength of unreinforced
masonry structures.
Keywords: Masonry wall, disturbed state concept, softening behavior, macro modeling, compressive behavior, tensile behav-
ior.

1. INTRODUCTION results from the SAM code, which was developed by the
University of Pavia, the Genoa research group and the
Masonry buildings are constructed in many parts of the
Basilicata research group [6]. The Basilicata research group
world where earthquakes occur. Hence, knowledge of their
used a no-tensile-strength macro-element model with crush-
seismic behavior is necessary to evaluate the seismic per-
ing and shear failures while the Genoa research group used a
formance of these types of buildings. Pushover analysis is
finite element model with layer failures. The ultimate base
commonly used to evaluate seismic performance and deter-
shear force for wall A was predicted to be 1682 kN by the
mine the capacity curve. Therefore, the capacity curve of
Genoa R.G., 1339 kN by SAP2000, 1115 kN by the SAM
unreinforced masonry buildings studied in this paper.
code, and 1395 kN by the Basilicata R.G. Accordingly, the
An approach for analysis of unreinforced masonry build- ultimate base shear force for wall B is predicted to be 650
ings is the macro-modeling of masonry as a composite mate- kN by the Genoa R.G., 474 kN by the SAM code and SAP
rial. The macro modeling is more practice oriented due to the 2000, and 508 kN by the Basilicata research group. These
reduced time and memory requirements as well as a user- results show differences between the different studies.
friendly mesh generation. The compressive strength of a Hence, they may lead to the confusion of practicing engi-
masonry unit is an important parameter in the analysis of neers, because they can not figure out which codes or re-
unreinforced masonry buildings using the macro-element search results are the most applicable or precise.
method. A masonry unit includes mortar joints and masonry
The in-plane shear behavior of hollow brick masonry
bricks. The compressive strength for masonry units with
panels was evaluated [7]. The nonlinear behavior of masonry
different mortar was evaluated in previous researches [1-5].
was modeled assuming elastic-perfectly plastic behavior,
SAP2000 v.10, a software package with a user-friendly Drucker-Prager, of the mortar joint in the ANSYS 5.4 com-
interface which is widely used by practicing engineers, was mercial software. In other words, the micro-element method
used for the seismic analyses of masonry buildings [6]. Two was used to analyze the panels. A comparison between the
unreinforced stone masonry walls in the Catania Project experimental results and numerical analysis shows good
were modeled with SAP2000 v.10. The static pushover agreement. A macro-element approach to the three-
curves from the analyses were compared with predicted dimensional seismic analysis of masonry buildings was ap-
plied [8]. The full model displays a base shear force which is
*Address correspondence to this author at the Department of Civil approximately 25% higher than the value calculated for the
Engineering, Faculty Engineering, Razi University, P.O. Box: 67149 plane structure. Seismic fragility of an unreinforced masonry
67346, Kermanshah, Iran; Tel.: +98 831 427 4535; Fax: +98 831 427 4542; low-rise building was studied using a structural modeling
E-mail: Ahakhaveissy@razi.ac.ir method. The method utilizes a simple, composite nonlinear

1874-1495/12 2012 Bentham Open

The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 201

spring. In this method, the wall is divided into distinct areas bending moment shear force strength domains of the ma-
or segments. Each segment of the unreinforced masonry wall sonry spandrels were derived by means of a heterogeneous
is then represented by a nonlinear spring, and the springs are upper-bound finite element limit analysis. The results were
assembled in series and in parallel to match the segment to- stored in a database. In step 2, an equivalent frame model of
pology of the wall [9]. Rota et al. [10] presented a new ana- the masonry wall was assembled. In the frame model, the
lytical approach for the derivation of fragility curves for ma- spandrels were modeled as elastic Timoshenko beams. At
sonry buildings. The methodology was based on nonlinear each analysis step, a check was performed to determine
stochastic analyses of building prototypes. Monte Carlo whether the internal forces of these coupling beams were
simulations were used to generate input variables from the smaller than the failure loads stored in the database created
probability density functions of mechanical parameters. in step 1. The shear force and bending moment capacity of
The seismic performance of existing unreinforced ma- the piers were simply estimated according to the Italian De-
sonry buildings in North America was considered in a state- sign code. The proposed analysis approach was appeared
of-the-art paper [11]. The various failure modes of unrein- capable of deriving the pushover curve of unreinforced ma-
forced masonry buildings subjected to earthquake excitation sonry walls [21]. A constitutive model was developed on the
were described in the paper. The damage to the existing basis of homogenized anisotropic elasto-plasticity for analy-
buildings for different earthquake scenarios was evaluated. sis of unreinforced masonry. The effect of anisotropy was
introduced by a fictitious isotropic stress and strain space.
The static pushover curve was studied using the bound- The advantage of this model is that the classical theory of
ary element method for unreinforced masonry walls [12]. In plasticity can be used to model nonlinear behavior in the
the analysis, a no-tension-material with an infinite strength in isotropic spaces [22].
compression was adopted to model the masonry buildings.
The predicted results showed good agreement with experi- A rigid-body numerical model was used to identify the
mental data. minimum height-to-thickness ratio that would cause the wall
to collapse when subjected to different out-of-plane ground
Milani et al. [13] performed a three-dimensional ho- motions [23]. The spectral accelerations of the ground mo-
mogenized limit analysis to determine the ultimate lateral tions were selected to be 0.24 g, 0.3 g, 0.37 g and 0.44 g. The
load of full masonry structures. Linearized homogenized model was calibrated using the results of full-scale shake
surfaces for masonry in six dimensions [14, 15] were ob- table tests of a wall with a height to thickness ratio of 12
tained and implemented in a finite element code. Compari- [24]. The results of the analysis showed that when a wall is
sons between the predicted results from the 3D homogenized subjected to a spectral acceleration of 0.44 g. The probability
limit analysis and experimental data show an error of ap- of collapse for height-to-thickness ratios less than 10 is less
proximately 12%. Milani et al. [16] also used a 3D homoge- than 1%. The ratios for spectral accelerations of 0.24 g and
nized limit analysis for full masonry buildings reinforced by 0.3 g are 18 and 15, respectively [23]. Therefore, walls with
FRP. The error between the predicted results and experimen- the conditions described above will be stabilized when sub-
tal data for a two-story masonry building is 4.6% in the ab- jected to out-of-plane ground motions. More over the in-
sence of FRP and 9.4% in the presence of FRP. Milani [17, plane strength of the wall will be important in resisting lat-
18] applied the 3D homogenized limit analysis to determine eral forces.
the limit load of a wall under in-plane and out-plane loading.
Akhaveissy [25] presented a new close form solution to
A simple equilibrium model was used to estimate the determine the shear strength of unreinforced masonry walls.
ultimate capacity of masonry shear walls. The model was Predicted results show less error percentage than ATC and
based on strut-and-tie schemes representing the combination FEMA-307 [26]. The new explicit formula is based on re-
of the compression or tension stress fields at the ultimate sults of proposed interface model by Akhaveissy [25]. Con-
condition. Comparisons between the performance of the sequently, the proposed closed form solution can be used
model and experimental results for dry-joint and mortar-joint satisfactorily to analyze unreinforced masonry structures.
masonry show good agreement. [19]
The research results related to macro-modeling processes
A finite element analysis conducted for a single-story, which discussed above show considerable differences be-
one-room masonry building, with different aspect ratios and tween various methods of macro-modeling in comparison
with different positions of wall openings, subjected to a with test data [6].
seismic force with varying direction [20]. The response spec-
The research described above on the numerical analysis
trum method was employed for the analysis. The predicted
of masonry walls shows that numerous researchers have
results show that the critical direction of seismic force for the
modeled the behavior of masonry walls using the micro-
development of maximum stresses in the walls of a room modeling approach which is suited for the analysis of small
occurs when the opening is along the short wall of the room. masonry walls. However, the macro-modeling approach can
It was also observed that the maximum principle tensile be used for the analysis of large practical problems. There-
stress occurred in the short wall, and the maximum shear fore, investigations into the use of macro-modeling, based on
stress occurred in the long wall. an appropriate constitutive law for the analysis of masonry
The analysis of unreinforced masonry buildings em- walls, are recommended and are the focus of this paper. In
ployed a two-step approach [21]. In step 1, the ultimate this paper, the disturbed state concept and hierarchical single
surface (DSC/HISS) model is modified to analyze masonry
202 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

structures in which different HISS single surfaces are used 2. DISTURBED STATE CONCEPT WITH HISS
for the compressive and tensile behaviors of the masonry. MODEL
The modified model is called DSC/HISS-CT, where CT de-
notes compression and tension. The base model (HISS) was The following description of the DSC model is adopted
introduced by [27-31]; a brief description is given below. from various publications, e.g., [27-31]. In this model, a de-
forming material element is assumed to be composed of two
The DSC/HISS model is a unified and hierarchical model (or more) reference states, the relatively intact (RI) and the
which can be used to characterize elastic, plastic and creep fully adjusted (FA) Fig (1). The material is assumed to be
deformations, as well as micro-cracking that leads to fractur- transformed continuously from the relative intact (RI) state
ing and failureing, degradating or softening, and healing or to the fully adjusted (FA) state, Fig (1a), at randomly dis-
strengthening. It has been used to model a wide range of tributed locations under external excitations such as me-
materials such as clays, sands, concrete, asphalt concrete, chanical and thermal loading. The transformation involves
ceramic, metals, alloys and silicon, as well as interfaces and micro-structural changes that cause particle reorientation and
joints [31]. It has been implemented in nonlinear finite ele- relative motions. The observed behavior is expressed in
ment procedures to solve a wide range of engineering prob- terms of the RI and FA states using the disturbance function,
lems including two- and three-dimensional [31-33] and cy- D, which acts as a coupling or interaction mechanism be-
clic loading [31, 34, 35] problems. Thus, the DSC model has tween the RI and FA states, Fig (1b). The disturbance grows
a number of advantages compared with other available mod- as the material deforms and the plastic strain (or work) ac-
els that often account for only specific behavioral aspect(s). cumulates. Thus, DSC intrinsically includes coupling in
The DSC/HISS models have also been used to model the which the micro-cracked (damaged) or fully adjusted part
behavior of concrete and rocks [28, 30, 31]. In this paper, the also contributes to the response of the material. The RI and
DSC model is used for unreinforced masonry structures. The FA states can be defined using various models. Continuum
new feature is the use of the HISS yield function for yield in elasticity or plasticity can be used to model the response of
both compression and tension. the RI state while the FA state can be assumed to carry only

(a)
RI

Observed
D
Residual

FA

(b)
Fig. (1). (a) RI and FA states in DSC, and (b) disturbance as a coupling between the RI and FA states.
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 203

(1a), the incremental stress-strain equations for the RI (plas-

ticity) model are derived as [31]:
T
Q F e
Ce C
d C e d
T
F
F e Q
(3)
C F

1/ 2
Q T Q
F

where is the stress vector, C e is the elastic constitutive

matrix, d denotes an increment, and Q is the plastic poten-
Fig. (2). HISS yield surface in J 1 J 2 D space.
tial function. When the associated flow rule is adopted,
F Q.
hydrostatic stress or can be modeled using the critical state
model [31]. Brief description of the models for the RI and The compressive stress is assumed to be positive for geo-
FA states and the disturbance used in the DSC model are logic materials and masonry. The yield surface, F is valid
given below. for compressive behavior in the positive J 1 J 2 D space,
Fig. (2). The behaviors of materials such as concrete and
2.1. Relative Intact (RI) State masonry are different under compression and tension; thus,
The hierarchical single-surface (HISS) plasticity model the yield surface, Fig. (2), is not appropriate on the negative
[27] provides a general formulation for the elasto-plastic J 1 -axis. A ad hoc model such as the stress transfer approach
characterization of material behavior. It involves a single [36], is often used. In this approach, the computed tensile
continuous yield surface compared with some previous mod- stress above the tensile strength is redistributed in the zone
els which involved multiple (discontinuous) yield surfaces of the problem. The HISS model can be used for both com-
that resulted in computational difficulties. This model which pression and tension if the material parameters are deter-
allows for isotropic and anisotropic hardening as well as mined from appropriate laboratory tests under compression
associated and non-associated plasticity characterizations can and tension. A model, called DSC/HISS-CT can account for
be used to represent material responses based on the contin- both compression and tension yield. The model was intro-
uum plasticity theory [27, 31]. In the HISS model, the RI duced in [33, 37, 38]. However, DSC/HISS-CT has not yet
state is usually defined by the associated plasticity; accord- been used for the unreinforced masonry practice buildings
ingly, the yield function, F, Fig. (2), is given by presented in this paper. Fig. (3) shows yield surfaces for
n 2 compression and tension. As it can be seen, although the
F J 2 D ( J 1 J 1 )( 1 S r )0.5 0 yield surfaces plotted in the same stress space for conven-
J 2D ience, are different from each other. Besides, their parame-
J 2D ters are determined from relevant compression or tension
pa 2
(1) tests. The yield surfaces in tension and compression are ex-
J 1 J 1 3 R / pa pressed in stress space in terms of J1, J2D and J3D, which in-
27 J 3 D clude all components of the stress vector in three-
Sr . dimensional space. A hierarchical single yield surface has
2 J 2 D 1.5
also been implemented in nonlinear finite element proce-

where J 2 D and J 3 D are the second and third invariants of

the deviatoric stress tensor, respectively. J1 is the first invari-
ant of the total stress tensor, pa is the atmospheric pressure,
R is the reference stress and is used mainly to include the
intercept ( c ) Which is proportional to the cohesive strength
Fig. (2). and are related to the ultimate condition. The
hardening or growth function for the plastic yield can be
expressed as
a1
(2)
1
Fig. (3). Schematic of compressive and tensile HISS yield surfaces
where a1 and 1 are the hardening parameters, and is the in J 1 - J 2 D Stress Space [37].
trajectory or the accumulated plastic strains. Using F, Eq.
204 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

dures to solve a wide range of two- and three-dimensional the corresponding values of D .
engineering problems [31-33]. Therefore, the model can be
used to analyze unreinforced masonry structures as 3D Akhaveissy and Desai introduced a critical disturbance to
boundary value problems and can include the out-of-plane show the failure zones [37, 38]. The value of the critical dis-
seismic effects. turbance, Dc, was assumed to occur at about 0.9. The critical
disturbance showed that micro-cracking may start when the
2.2. Fully Adjusted (FA) State disturbance is about 0.003 for compression and about 0.0001
for tension. Micro-cracks then grow and coalesce into cracks
Various characterizations of the behavior of the FA state that lead to fracture or failure at the critical disturbance, Dc =
are given in Desai [31]. In a simple form, the FA state for 0.9 [37, 38]. Thus, D can be used as a measure to identify the
masonry can be considered using residual stress in the stress- initiation and growth of micro-cracks (based on test data)
strain curve response, Fig. (1b). Fully adjusted stress can that lead to fracture and failure. For example, when the dis-
also be based on approximately 20% of the compressive re- turbance reaches 0.9 or higher values, fractures occur and
sistance of masonry in uniaxial tests [2], and the residual grow. In finite element analysis, elements that reach this
stress has been considered as 14% of the compressive resis- critical value are identified after each load increment; thus,
tance of masonry in uniaxial tests [39]. the initiation and growth of fractures are provided progres-
Disturbance sively by the computer procedure. In later applications, the
contours of D are plotted based on the computational results,
Disturbance, D , can be defined in terms of measured
and the cracking and fracture are related to the values and
stress, void ratio, pore water pressure, and nondestructive
extent of the disturbance.
properties, such as P or S wave velocities [31]. In terms of
stress, it is defined as: Both the RI and FA states contribute to the material re-
sponse through disturbance (D) as the coupling function. The
i a
D (4) following DSC equations in incremental form show this
i c coupling [31]:

where i , a and c are RI, observed and FA stress values, d ija ( 1 D )d iji Dd ijc dD( ijc iji ) (6)
respectively. To introduce D in the DSC model, Eq. (6), it
needs to be expressed in a mathematical form in terms of
where d denotes an infinitesimal increment, and ij is the
basic variables, such as accumulated plastic strains or work. stress tensor. The DSC model in this study is based on small
Hence, D is expressed in terms of the accumulated deviatoric strains. It can be modified by including large strains.
plastic strains using the Weibull [40] type function:
2.3. Numerical Simulation of Masonry Prism Behavior
D Du [ 1 exp( A DZ )] (5) Laboratory Tests (Fig. 5)
where Du is the ultimate disturbance (often assumed to be On
unity), D is the trajectory of (deviatoric) plastic strains, and the basis of a number of tests on masonry the following
mathematical equations have been used by Kaushik et al.
A and Z are disturbance parameters. The parameters in Eq.
[1, 2] and Pandey and Meguro [39]:
(5) are determined on the basis of the values of the measured
a at various points on the stress-strain curve, Eq. (4), and

fc ft
f m' f t f m'

0.85 f m'

0.2 f m'

0 cu 1 t m

(a) (b)
Fig. (4). (a) Compressive and (b) tensile behavior of a masonry prism.
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 205

For Compressive Behavior: sumed to determine the parameters of DSC for the softening
behavior of masonry in tension. However, the DSC/HISS-
2 CT model accounts for the nonlinear behavior of masonry in
f m f m' 2 0
0 0 the tensile region in terms of an exponential strain function

(Eq. 5 and Fig 11). In Eq. 7, f m' is the compressive strength
0
f m f m' 1 0.15
0 1 (7 a,b,c) of a masonry prism,

cu 0

0.002 0 0.004 , 0.003 cu 0.007 [1-2], m 10 t ,
f m 0.2 f m' 1
varies between 0.1 to 0.25, E c is the modulus of elastic-
Eq. 7(a) is Hognestads model for concretes compressive
behavior [41]. Eq. 7(b) is a linear function between peak
ity of masonry, and 1 , t are shown in Fig. (4). Fig. (4a)
stress and residual stress. Eq. 7(c) is the residual stress that is shows schematic plots of Eqs. 7(a) to 7(c) for compression,
assumed to be 20% of the peak stress. and Fig. 4(b) shows the plot for tension.
For Tensile Behavior: Eqs. 7(a) to 7(c) are applied to construct the compressive
behavior of masonry prisms. Similarly Eqs. 7(d) and 7(e) are
f t Ec t applied to construct for the tensile behavior. In this sense, the
t stress-strain behavior used here is called constructed be-
f t f m' 1 t havior.
m t (7 d,e)
2.4. Determination of Parameters.
Eq. 7(d) is obtained from the elastic behavior of masonry
in the tensile region up to the peak tensile stress. Softening The DSC/HISS-CT parameters are determined based on
behavior occurs after this point. This behavior is modeled by Eqs. 7(a) to 7(c) for the compressive behavior of masonry
a linear function. Here, a linear function (Eq. 7(e)) is as- prisms and Eqs. 7(d) and 7(e) for the tensile behavior. The

Table 1. Parameters for the DSC/HISS-CT Model

Compressive strength

DSC Parameters (Distur-

HISS plasticity parameters
Kind of behavior
Kind of material

bance)

E n R a1 1 A Z
7635.0 (MPa)
Compression

6.6629e11
0.3445

5.0846
0.5e-9
0.335

1.900
0.15

0.30

5.23
Masonry wall

10.50 (MPa)

7635.0 (MPa)

27065.09
Tension

0.4000

1.3308
1.0e-9
0.030

1.000
0.15

0.40

1.25
Compression

1750 (MPa)

4865.866
0.3000
1.2e-9
0.335

1.893
5.50

1.70
0.2

0.3
The Pavia Test

7.90 (Mpa)

1750 (MPa)
Tension

35000
0.030

1.100
7e-6
0.2

0.4

0.5

0.2

0.4
206 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

details on how to determine the parameters are given by are available. In addition, a computer code that allows the
Desai [31], Akhaveissy and Desai [37, 38] and Akhaveissy et use of the HISS yield surface for both compression and ten-
al. [42]. Table 1 shows the parameters of the model for the sion in masonry has been developed [42]. The iterative-
compressive behavior of masonry prisms with compressive incremental method with an initial stiffness scheme was ap-
strengths f m' = 10.5 and 7.9 MPa, respectively. These values plied to analyze unreinforced masonry structures. This
method yielded accurate and convergent results for the prob-
for f m' are chosen to be consistent with masonry prisms in lems. However, the method can not predict the softening
later applications. Akhaveissy and Desai [37] determined the behavior of the displacement load curve of structure.
parameters of the model for the masonry prism with com- Hence, the method can be used to determine the linear and
pressive strength f m' = 10.5 MPa. hardening behavior of the curve. It is noticeable that parts of
unreinforced masonry wall can show the softening behavior
[37]. The results of the nonlinear computer analyses are
2.5. Specimen Level Validations
compared with the observed data below.
The model is validated at the element level by using the
parameters in Table 1 and Eqs. (6) and (3a). At the specimen 3.1. A Masonry Shear Wall
level, predictions are obtained by integrating the incremental
constitutive relations, Eq. (3a). Akhaveissy and Desai [37] The mechanical response of the masonry wall illustrated
compare predicted results of the stress-strain curve from the in Fig. (6) was simulated. Experimental tests on a masonry
wall were carried out by Raijmakers and Vermeltfoort [43].
model for the masonry prism with compressive strengths f m' With regard to specimens with and without openings two
= 10.5 MPa and the constructed stress-strain curves by Eq. 7. series were tested. Here only windowed panel is considered
Hence, the stress-strain curve obtained by the model the con- for the sake of conciseness. The wall was made of wire-cut
structed curve are evaluated for the masonry prism with solid clay bricks with dimensions of 210 mm x 52 mm x 100
compressive strength f m' = 7.90 MPa. Fig. (5) compares mm and 10 mm thick mortar joints. The height/width ratio
predictions from the model of the masonry prism with com- was 1. The dimensions of the wall were 1000 mm x 990 mm.
Two stiff steel beams at the horizontal boundaries of the test
pressive strengths f m' = 7.9 MPa to the constructed stress- setup clamped the top and bottom walls. The geometry of the
strain curves by Eq. 7. model and the boundary conditions are shown in Fig. (6).
It is clear from Fig. (5) that the results from the The homogenization model was implemented to analyze the
DSC/HISS-CT models match well with the observed behav- behavior of the masonry wall [17, 18]. It showed that the
ior based on the constructed compressive and tensile behav- responses of the numerical model matched the experimental
iors with different peak stresses, different softening region data. However, the masonry wall was analyzed with the
slopes and different residual stresses, Fig. (5). DSC/HISS-CT model using the macro-modeling method.
Bricks and mortar joints are represented by 8-node plane
3. APPLICATIONS stress continuum elements for the numerical analysis. The
Numerical simulations of masonry structures and load- wall is modeled by 186 eight-noded isoparametric elements;
displacement responses are considered for masonry prisms the mesh has 1230 DOF. The value of the error for the dis-
with different compressive strengths. Computer programs for placement convergence criterion was 5e-4. Fig. (7) shows
two-dimensional analyses [34] with the DSC/HISS model the mesh that was used for the analysis of the wall and the

9 0.08
8 Constructed
Compressive stress

0.07
Tensile stress

7
0.06
6 DSC/HISS-CT
0.05
5
0.04
4
3 0.03
2 0.02
1 0.01
0 0
0 0.01 0.02 0.03 0 0.0002 0.0004 0.0006

Axial strain Axial strain

(a) (b)

Fig. (5). Comparison between predictions and data for (a) compressive and (b) tensile behavior of a masonry prism in Pavia test.
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 207

q= 30 N/cm2 in this paper is within the range of reported values.

After point A in Fig. (8), the analysis in the model did
F
not converge. Fig. (8) shows that the predicted ultimate load
Stiff Steel is about 46.8 kN, compared with 47 kN from the test data, or
Beam an error of about 0.4%. In addition, the disturbance parame-
ter for point A from Fig. (8) is shown in Fig. (9). A signifi-
100 cm cant part of the wall reached the critical disturbance around
0.90. This result shows the failure of the wall. The maximum
disturbance is 1.00. The predicted failure pattern in Fig. (9)
shows that the collapse for both piers was occurred at dis-
placement equal to 14.2 mm, point A in Fig. (8). It is notice-
able that the value of the disturbance parameter for both
yield surfaces in tension and compression is changed be-
99 cm
tween zero to one. Hence, it is not possible to distinguish
among shear, tensile and compressive cracking when the
Fig. (6). Geometry of the masonry wall (dimensions in [cm]). disturbance parameter is plotted. Therefore, the length of the
tensile crack for piers may be evaluated as equal to the width
boundary conditions. of the piers. Hence, in the following, the lateral force is esti-
mated only based on the fracture length and the calibrated
Fig. (8) shows a comparison between the load displace-
parameters that were obtained from the stress level analysis,
ment curve predicted by the DSC/HISS-CT model and the
Table 1. The cohesive strength for tensile behavior, Fig. (3),
test data. The agreement between the predictions and the test
and the estimate of the lateral load for each pier based on
data is considered to be good. The analysis of the wall by
proposed explicit formula [37] are
DSC/HISS-CT did not converge after point A in Fig. (8).
The initial slope of the predicted force-displacement Ct 3 Rt t 3* 1.0* 0.03 0.52
curve from Fig. (8) agrees with the experimental data. It T Ct * Lt * t 0.52 * ( 390 )* 100 20280 N
shows that the predicted elasticity modulus is essentially the C P T 20280 0.3 * 100 * 390 31980 N
same as that from the test data. The elasticity modulus for a F1 t * T c * C 0.03 * 20280 0.335 * 31980 22022 N
compressive strength of 10.5 MPa is 7635 MPa, Table 1,
which is 727 times f m' ; the ratio given in FEMA 306 [45] is Hence, the lateral resistant forced is for both piers as
550. The International Building Code [46] and MJSC docu- F 2 F1 2 * 22022 44044 N 44.1 kN
ments [47] recommend E to be 700 times f m' , and the Cana-
In the above calculation, t is the thickness and Lt is
dian masonry code S304.1 [48] recommends E to be 850
named the length of the tensile crack. The estimated lateral
times f m' . Thus, the predicted ratio using the model proposed load is about 44.1 kN, compared with 47 kN from the test
q= 30 N/cm2
F ,
100 cm

99 cm Fixed node

Fig. (7). The boundary conditions and mesh used for the analysis of the masonry wall.
208 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

50 Point A
45
40
Lateral Load (kN) 35 Point B
30
Point C
25
20 Present work by DSC/HISS-CT model
15 Experimental [43]
10 Heterogeneous approach [44]
5 Limit Analysis [18]
0
0.00 1.25 2.50 3.75 5.00 6.25 7.50 8.75 10.00 11.25 12.50 13.75 15.00

Top Displacement (mm)

Fig. (8). Comparison between the test data and the predicted load-displacement curve for the windowed masonry wall.

Failure region

Fig. (9). The variation of disturbance parameter for point A in Fig. (8).

data, or an error of about 6.2%. Therefore, the resistant lat- and A, Fig. (10b) and (c), show the plastic deformation for
eral load may be estimated based on the length of the tensile the wall. Fig. (10c) shows that the bottom of the left pier is
crack. collapsed due to combination of the shear stress and the ten-
sile stress. Likewise the right pier of the wall is collapsed
The predicted failure pattern in Fig. (9) shows that the
due to combination of the shear stress and the compressive
wall is collapsed under the lateral load equal to 46.8 kN,
stress. The oblique vector on Fig. (10c) shows the direction
Point A in Fig. (8). Hence, successive predicted deformed
of the deformation for the right pier. The deformation of the
shapes of the wall are evaluated in Fig. (10). The deformed
right pier implies the shear failure mode for the pier. It is
shapes are in accordance with points A, B and C in Fig. (8).
noticeable that the corners of the window, Fig. (10c), are
Fig. (10a) shows the deformation of the wall for dis- deformed due to the lateral load.
placement equal to 0.54 mm, point C in Fig. (8). It should be
noted that the point C is on linear part of the curve. Hence, 3.2. A Two-bay, Two-story Building
the plastic deformation should not be observed when the wall
A full-scale, two-story unreinforced masonry building
is subjected to the lateral load of the point C. Fig. (10a)
tested at Pavia University was chosen for model validation
shows the small deformation of the wall for the lateral load
[49]. This structure has been extensively investigated in the
equal to the point C. However, deformed shapes for points B
literature [50-52]. The building, with a 6*4.4 m floor plan
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 209

(a) Deformed shape for point (b) Deformed shape for point (c) Deformed shape for point A
C in Fig. 8 equal to 0.54 mm B in Fig. 8 equal to 1.80 mm in Fig. 8 equal to 14.2 mm
Fig. (10). Deformed shapes of the wall for different levels of the lateral displacement; the displacements are exaggerated by a factor of 5.

and 6.4 m in height, contains an almost independent shear DSC/HISS-CT model were determined in Table 1. Calderini
wall that is in-plane loaded. The wall considered here et al. [51] used the finite element method (FEM) to analyze
(named the door wall) is 250 mm thick and has two doors the two-story unreinforced masonry building tested at Pavia
on the first story and two windows on second story, as University. The model included 2696 nodes and 5128 trian-
shown in Fig. (11). The door wall includes two exterior piers gular shell elements. The used mesh in present analysis by
and one interior pier. The exterior pier width and axial loads DSC/HISS-CT model is shown in Fig. (12). The mesh in-
on the bottom and top levels are equal to 1.15 m, 56 kN and cluded 1862 nodes and 496 eight-noded isoparametric ele-
26.9 kN, respectively. The interior pier width and axial loads ments. Fig. (13) shows comparisons between the present
on the bottom and top levels are equal to 1.82 m, 133 kN and work and experimental data and numerical analyses [51].
64.5 kN, respectively.
The predicted results were compared with results ob-
The properties of the structure used in the model are tained using the equivalent frame model and the Tremuri
summarized below [50]: software [51]. The equivalent frame included 9 nodes, and 3
The maximum compressive strength of a masonry prism, nodes were fully constrained at the base. The reduce stiffness
fm, is equal to 7.9 MPa. The joint tensile strength and the and full stiffness were used to analyze the building with the
joint cohesion are 0.07 MPa and 0.14 MPa, respectively. The equivalent frame. Akhaveissy [52] also analyzed the door
joint coefficient of friction is 0.55. The shear modulus is wall by the equivalent frame based on proposed close form
equal to the effective value, Geff = 90fm. The parameters for solution [25].

2F

Opening

Equivalent
frame
1.15 m 1.82 m 1.15 m
Exterior pier Interior pier Exterior pier

Fig. (11). Door wall of the full-scale, two-story unreinforced masonry building tested at Pavia University.
210 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

point A, is evaluated in Fig. (14). It is noted that a distur-

bance parameter greater than 0.9 indicates failure. Crack
patterns from the experimental test of the URM building at
the failure state (a top displacement equal to 24 mm) show
damage to the piers for the second story as sliding failure and
the first story as diagonal crack as well as damage to the
deep beam between doors and windows at the first floor [50].
The predicted failure of the piers for the second story and the
predicted failure of the deep beams for both stories correlate
with the observed data while the damage to the piers for the
first story is not seen in the present work. It is noticed that
the local failure occurred at the bottom of the piers for the
first story. This difference is due to the dissipation of energy
by the piers and the deep beams of the second story. Hence,
the second story is as the instable floor at the displacement
equal to 31.7 mm and the convergency of the analysis is not
occurred for the base shear force larger than 154 kN. The
Fig. (12). The mesh used for the analysis of the two-story unrein-
forced masonry building tested at Pavia University. deformed shape for point A in Fig. (13) is also plotted in Fig.
(15). It should be noticed that form of deformation at the
right corner of the roof is due to the concentration of applied
Fig. (13) shows that the equivalent framereduce stiff- lateral external load equal to 2F , Fig. (11).
ness model predicted that the stiffness of the building would
be lower than the real stiffness. The value of the ultimate The base shear force is estimated below based only on
base shear force from the Tremuri software was estimated to the explicit formula [37] and calibrated parameters that were
be 167 kN whereas the experimental value was determined obtained from the stress level analysis, Table 1. The strength
to be 147 kN. The value of ultimate base shear force from lateral force is calculated for each of the piers, and the base
the present work and from Belmouden and Lestuzzi [50] was shear force is estimated by minimum the sum of the strength
predicted to be 154 kN and 138.4 kN, respectively. The finite forces of the piers for the first and second story. The strength
element method estimate of the force was 157 kN. Fig. (13) lateral force is shown in Table 2. To determine the strength
shows a better agreement between the test data and the pre- of the pier, the fracture length is assumed to be equal to the
sent work than with equivalent frame models and finite ele- width of the pier [37]. Therefore, the cohesive strength for
ment method. To consider the damage to the structure, the tensile behavior, Fig (3), and the estimate of the lateral load
disturbance parameter for the roof displacement in Fig. (13), for the interior pier (Fi) and the exterior pier (Fe), are as fol-

180

160 Point A
140
Base Shear force (kN)

120

100
DSC/HISS-CT
80 Experimental [50]
FEM [51]
60
Equivalent Frame - Full Stiffness [50]
40 Equivalent Frame - Reduce Stiffness [50]
Belmouden and Lestuzzi [50]
20
Akhaveissy [52]
0
0 5 10 15 20 25 30 35 40
Displacement at Roof (mm)

Fig. (13). Comparison of the predicted results with experimental data for the two-story unreinforced masonry building tested at Pavia Uni-
versity.
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 211

Fig. (14). Damage levels for the URM structure from the present work for displacement at the roof equals 31.7 mm, point A in Fig. (13).

Failure
region Failure
Failure Failure
region
region region

Failure
Failure Failure
Failure
ion
regregion gion
reregion

Fig. (15). Deformed shape of the wall for the base shear force of the point A in Fig. 13; the displacements are exaggerated by a factor of 2

lows. Subscript i is shown for the interior pier and e for the
C t 3 Rt t 3* 0 .2 * 0 .03 0 .10392
exterior pier.
212 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

Table 2. The Ultimate Shear Force, F, for the Pavia Test

Story Pier T(N) P(N) C(N) F(N)

interior 47284 197500 244784 149868

1
exterior 29276 82900 112176 69478

The strength shear force for the first story 288823

interior 47284 64500 111784 72889

2
exterior 29876 26900 56776 38036

The strength shear force for the second story 148960

It is noticeable that the strength shear force for the stories CONCLUSIONS
are calculated based on the sum of the strength forces for
A nonlinear finite element method with eight-noded
both exterior piers and interior pier, Table 2. Hence, com-
isoparametric quadrilateral elements for combined masonry
parison between the strength shear force for the first story
blocks and mortar joints was used to predict the behavior of
and the second story show that the second story is collapsed
unreinforced masonry structures. The disturbed state concept
sooner than the first story. Consequently, the ultimate shear
(DSC) and the modified hierarchical single surface (HISS-
force is equal to the strength shear force of the second story.
CT) plasticity model with associated flow rules were used to
The estimated lateral load based on the close form solution is
characterize the compressive and tensile yields of the ma-
about 148.96 kN, compared with 147 kN from the test data,
sonry structures. The model can account for micro-cracking
an error of about 1.3%. Therefore, the resistant lateral load
in the masonry which leads to softening and fracture. The
may be estimated based on the explicit formulation presented
HISS-CT plasticity model involves two separate yield sur-
here.
faces for compressive and tensile yield; the continuous na-
The pushover curve in Fig. (13) provides also important ture of the proposed yield surface prevent computational
information on the ultimate drift of the walls, which is useful difficulties in currently available discontinuous or multiple
to determine the pseudo ductility. In particular, the ultimate surfaces models, such as critical state and cap models [31].
drift, u , is evaluated as the drift in relation of a shear The DSC model allows for the generation of discontinuities
in the material microstructure during loading (unloading),
strength degradation. The yield drift, y , is determined and it does not require external enrichments to allow cou-
equating the areas under the equivalent bilinear 1DOF model pling between continuous and discontinuous parts within the
and the actual capacity curve, as suggested by many codes of deforming material [55]. Moreover, the same framework in
practice [53]. In the present case values of the ultimate and the DSC can be used to model the behavior of interfaces and
yield drifts are equal to 31.7 mm and 3.64 mm, respectively. joints [31].
The pseudo ductility is hence determined as the ratio be- The DSC/HISS-CT model for masonry was validated at
tween ultimate and yield drift as: the specimen level. It was also applied successfully for a
u number of unreinforced masonry prism specimens. The
(8) computational predictions correlated well with test data and
y
the constructed model. The model was also implemented in a
From simulations results, the pseudo ductility for the finite element procedure to analyze the boundary problems.
The model predictions for the masonry shear wall and the
building is 8.7. The deflection amplification factor, R , can
unreinforced masonry building with two stories correlated
be determined through Newmark and Hall method [54] and well with the test data. The model can predict the hardening
is equal to 4.05. The over strength factor is also computed. It and softening behavior of materials. In addition, a closed
is equal to 1.85. The strength reduction factor due to the form solution was proposed based on the calibrated parame-
nonlinear hysteretic behavior is defined as the ratio between ters of the DSC/HISS-CT model. The closed form solution
the elastic and inelastic strength demand, it is equal to 0.29. predicts the ultimate lateral load of an unreinforced masonry
Finally, the seismic modification factor is determined as wall relatively well. Therefore, the DSC/HISS-CT model can
the product between the deflection amplification factor, the be used satisfactorily to analyze masonry structures similar
over strength factor, and the strength reduction factor. Con- to those considered herein.
sequently, in the present case, the seismic modification fac-
tor is equal to 2.17 which is comparable with experimental DISCLOSURE
data. Part of information included in this article has been pre-
viously published in ARCHIVES OF COMPUTATIONAL
The DSC Model for the Nonlinear Analysis of In-plane Loaded Masonry Structures The Open Civil Engineering Journal, 2012, Volume 6 213

METHODS IN ENGINEERING Volume 18, Number [19] P. Roca, Assessment of masonry shear-walls by simple
equilibrium models, Constr. Build. Mater., vol. 20, pp. 229-238,
4 (2011), 485-502, DOI: 10.1007/s11831-011-9067-4. 2006.
[20] M. Shariq, H. Abbas, H. Irtaza, and M. Qamaruddin, Influence of
CONFLICT OF INTEREST openings on seismic performance of masonry building walls,
Build. Environ., vol. 43, pp. 1232-1240, 2008.
The authors confirm that this article content has no con- [21] G. Milani, K. Beyer, and A. Dazio, Upper bound limit analysis of
flicts of interest. meso-mechanical spandrel models for the pushover analysis of 2D
masonry frames, Eng. Struct., vol. 31, pp. 2696-2710, 2009.
[22] J. Lopez, S. Oller, E. Onate, and J. Lubliner, A homogeneous
ACKNOWLEDGEMENTS constitutive model for masonry, Int. J. Numer. Meth. Eng., Vol.
46, pp. 1651-1671, 1999.
None decleared. [23] I. Sharif, C.S. Meisl, and K.J. Elwood, Assessment of ASCE 41
Height-to-thickness ratio limits for URM walls, Earthquake
REFERENCES Spectr, vol.23, no. 4, pp. 893-908, 2007.
[24] C.S. Meisl K.J. Elwood, and C.E. Ventura, Shake table tests on
[1] H.B. Kaushik, D.C. Rai, and S.K. Jain, Uniaxial compressive the out-of-plane response of unreinforced masonry walls, Can. J.
stress-strain model for clay brick masonry Curr. Sci., vol. 92, no. Civ. Eng., vol. 34, pp. 1381-1392, 2007.
4, pp. 497-501, 2007. [25] A.H. Akhaveissy, Lateral Strength Force of URM Structures
[2] H.B. Kaushik, D.C. Rai, and S.K. Jain, Stress-Strain based on a Constitutive Model for Interface Element, Lat. Am. J.
Characteristics of Clay Brick Masonry under Uniaxial Solids Struct., vo. 8, pp. 445-461, 2011.
Compression, J. Mater. Civil Eng. ASCE, vol. 19, no. 9, pp. 728- [26] Applied Technology Council (ATC), Evaluation of earthquake
739, 2007. damaged concrete and masonry wall buildings, Technical
[3] A. Tena-Colunga, A. Juarez-Angeles, and V.H. Salinas-Vallejo, Resources Publication no. 307, Federal Emergency Management
Cyclic behavior of combined and confined masonry walls, Eng. Agency (FEMA-307), Washington (DC), 1999.
Struct., vol. 31, pp. 240-259, 2009. [27] C.S. Desai, S. Somasundaram , and G. Frantziskonis, A
[4] K. Chaimoon, and M.M. Attard, Experimenatl and numerical hierarchical approach for constitutive modeling of geologic
investigation of masonry under three-point bending (in-plane), materials, Int. J. Numer. Anal. Methods Geomech., vol. 10, pp.
Eng. Struct., vol. 31, pp. 103-112, 2009. 225-257, 1986.
[5] G. Vasconcelos, and P.B. Lourenco, Experimental [28] C.S. Desai, and M.R. Salami, A constitutive model for rocks, J.
characterization of stone masonry in shear and compression, Geotech. Eng., ASCE, vol. 113, pp. 407-423, 1987.
Constr. Build. Mater., vol. 23, pp. 3337-3345, 2009. [29] C.S. Desai, and J. Toth, Disturbed state constitutive modelling
[6] L. Pasticier, C. Amadio, and M. Fragiacomo, Non-linear seismic based on stress-strain and nondestructive behavior, Int. J. Solids
analysis and vulnerability evaluation of a masonry building by Struct., vol. 33, pp. 1619-1650, 1996.
means of the SAP2000 V.10 code, Earthquake Eng. Struct. Dyn., [30] M.R. Salami, and C.S. Desai, Constitutive modeling including
vol. 37, pp. 467-485, 2008. multiaxial testing for plain concrete under low confining pressure,
[7] A. Gabor, E. Ferrier, and H.P. Jacquelin, Analysis and modeling J. Mater. Am. Concrete Inst., vol. 87, no. 3, pp. 228-236, 1990.
of the in-plane shear behavior of hollow brick masonry panels, [31] C.S. Desai, Mechanics of materials and interfaces, the disturbed
Constr. Build. Mat., vol. 20, pp. 308-321,2006. state concept, CRC Press: Boca Raton, Florida, U.S.A, 2001.
[8] A. Brencich, L. Gambarotta, and S. Lagomarsino, A macroelement [32] C.S. Desai, Mechanistic pavement analysis and design using
approach to the three-domensional seismic analysis of masonry unified material and computer models,Keynote Paper, Proc. 3rd
buildings. 11th European Conference on Earthquake Engineering, Intl. Symposium on 3D Finite Elements for Pavement Analysis,
Balkema, Rotterdam, 1998. Design and Research, Amsterdam. Netherlands, 2002.
[9] J. Park, P. Towashiraporn, J.I. Craig, and B.J. Goodno, Seismic [33] C.S. Desai, Unified DSC constitutive model for pavement
fragility analysis of low-rise unreinforced masonry structures, vol. materials with numerical implementation, Int. J. Geomech. ASCE,
31, pp. 125-137, 2009. vol. 7, no. 2, pp. 83-101, 2007.
[10] M. Rota, A. Penna, and G. Magenes, A methodology for deriving [34] C. Shao, and C.S. Desai, Implementation of DSC model and
analytical fragility curves for masonry buildings based on application for analysis of filed pile tests under cyclic loading, Int.
stochastic nonlinear analyses, Eng. Struct., vol. 32, pp. 1312-1323, J. Num. Anal. Method Geoemch., vol. 24, no. 6, pp. 601-624, 2000.
2010. [35] S.K. Pradhan, and C.S. Desai, DSC model for soil and interface
[11] M. Bruneau, State-of-the-art report on seismic performance of including liquefaction and prediction of centrifuge test, J.
unreinforced masonry buildings, J. Struct. Eng., vol. 120, no. 1, Geotech. Geoenviron. Eng. ASCE, vol. 132, no. 2, pp. 214-222,
pp. 230-251, 1994. 2006.
[12] C. Alessandri, and C.A. Brebbia, Strength of masonry walls under [36] O.C. Zienkiewicz, S. Valliapan, and I.P. King, Stress analysis of
static horizontal loads: boundary element analysis and experimental rock as a no tension material, Geotechnique, vol. 18, no. 1 , 1968.
tests, Eng. Anal., vol. 4, no. 3, pp. 118-134, 1987. [37] A.H. Akhaveissy, and C.S. Desai, Unreinforced Masonry Walls:
[13] G. Milani, P. Lourenco, and A. Tralli, 3D homogenized limit Nonlinear Finite Element Analysis with a Unified Constitutive
analysis of masonry buildings under horizontal loads, Eng. Struct., Model, Arch. Computat. Methods Eng., vol.18, no. 4, 485-502,
vol. 29, pp. 3134-3148, 2007. 2011.
[14] G. Milani, P. Lourenco, and A. Tralli, Homogenised limit analysis [38] A.H. Akhaveissy, and C.S. Desai, Application of DSC Model for
of masonry walls, Part I: Failure surfaces, Comp. Struct., vol. 84, Nonlinear Analysis of Reinforced Concrete Frames, Finite Ele.
pp. 166-180, 2006. Anal. Design J., vol. 50, pp. 98-107, 2012.
[15] A. Cecchi, and G. Milani, A kinematic FE limit analysis model fot [39] B.H. Pandey, and K. Meguro, Simulation of Brick masonry wall
thick English bond masonry walls, Int. J. Solids Struct., vol. 45, behavior under in plane lateral loading using applied element
pp. 1302-1331, 2008. method, 13th World Conference on Earthquake Engineering,
[16] G. Milani, E. Miani, and A. Tralli, Approximate limit analysis of Vancouver, B.C., Canada, August 1-6, 2004, paper No. 1664.
full scale FRP-reinforced masonry buildings through a 3D [40] W.A. Weibull, A statistical distribution function of wide
homogenized FE package, Compos. Struct., vol. 92, pp. 918-935, applicability, App. Mech., vol. 18, pp. 293-297, 1951.
2010. [41] C. Yalcin, and M. Saatcioglu, Inelastic analysis of reinforced
[17] G. Milani, Simple lower bound limit analysis homogenization concrete columns, Comp. Struct., vol. 77, pp. 539-555, 2000.
model for in-and out-of-plane loaded masonry walls, Constr. [42] A.H. Akhaveissy, and C.S. Desai, S.A. Sadrnejad, and H. Shakib,
Build. Mater., vol. 25, pp. 4426-4443, 2011. Implementation and Comparison of Generalized Plasticity and
[18] G. Milani, Simple homogenization model for the non-linear Disturbed State Concept for Load-Deformation Behavior of
analysis of in-plane loaded masonry walls., Comput. Struct., vol. Foundation, Scientia Iranica J., Trans. A: Civ Eng., vol. 16, no. 3,
89, pp. 1586-1601, 2011. pp. 189-198, 2009.
214 The Open Civil Engineering Journal, 2012, Volume 6 A.H. Akhaveissy

[43] T.M.J. Raijmakers, and A. Vermeltfoort, Deformation controlled [50] Y. Belmouden, and P. Lestuzzi, An equivalent frame model for
tests in masonry shear wall (in Dutch) Report B-92-1156, TNO- seismic analysis of masonry and reinforced concrete buildings,
Bouw, Delft, The Netherlands., 1992. Construct. Build. Mater. vol. 23, pp. 40-53, 2009.
[44] P.B. Lourenco, R. de Borst, and J.G. Rots, A plane stress [51] C. Calderini, S. Cattari, and S. Lagomarsino, In plane seismic
softening plasticity model for orthotropic materials, Int. J. Numer. response of unreinforced masonry walls: comparison between
Methods Eng., vol. 40, pp. 4033-57, 1997. detailed and equivalent frame models, COMPDYN 2009,
[45] Federal Emergency Management Agency (FEMA), Evaluation of ECCOMAS Thematic Conference on Computational Methods in
earthquake damage concrete and masonry wall buildings, basic Structural Dynamics and Earthquake Engineering In: Papadrakakis
procedure manual, ATC-43, FEMA306, Applied Technology M, Lagaros ND, Fragiadakis M (Eds.) Rhodes, Grecce, 22-24 June,
Council, Calif., 1999. 2009.
[46] International Code Council, International building code, 2003. [52] A.H. Akhaveissy, Finite Element Nonlinear Analysis of High-Rise
[47] Masonry Standard Joint Committee (MSJC), Building code Unreinforced Masonry Building, Lat. Am. J. Solids Struct., vol. 9,
requirements for masonry structures, ACI 530-02/ASCE 5- pp. 547-567, 2012.
02/TMS 402-2, American Concrete Institute, Structural [53] ATC40, Seismic evaluation and retrofit of concrete buildings,
Engineering Institute of the American Society of Civil Engineers., Vol. 1, Applied Technology Council (ATC), 1996.
The Masonry Society, Detroit., 2002. [54] N.M. Newmark, and W.J. Hall, Seismic design criteria for
[48] Canadian Standard Association (CSA), Design of masonry nuclear reactor facilities, Report No. 46, Building Practices for
structures, S304.1, Ontario, Canada., 2004. Disaster Mitigation, National Bureau of Standards, U.S.
[49] G. Magenes, and G.M. Calvi, In-plane seismic Response of Brick Department of Commerce, 209-236, 1973.
Masonry Walls, Earthquake Eng. Struct. Dyn., vol.26, pp. 1091- [55] H.B. Muhlhaus, Continuum Models for Materials with Micros-
1112, 1997. tructure, John Wiley & Sons: Chichester, U.K. 1995.

Received: February 08, 2012 Revised: July 10, 2012 Accepted: July 11, 2012

A.H. Akhaveissy; Licensee Bentham Open.

This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/
by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.