Performance Evaluation of a Reinforced Masonry

Model and an Unreinforced Masonry Model

Using a Shake Table Testing Facility
Shermi Chellappa 1 and R. N. Dubey 2
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.

Abstract: This paper presents behavior of both unreinforced masonry and reinforced masonry (RM) building models subjected to dynamic
loading with the help of shake table testing facility. The first masonry model was a half-scale conventional unreinforced masonry model. The
second model was strengthened with welded wire mesh (WWM) and coarse sand mortar. Both the models were subjected to several ground
motions. It was observed that the strengthened model performed much better than the unstrengthened model and there was no damage in the
strengthened model. In this paper, the observation of the test results and the influence of strengthening in terms of damage intensity and failure
pattern have been presented. DOI: 10.1061/(ASCE)CF.1943-5509.0001119. © 2017 American Society of Civil Engineers.
Author keywords: Unreinforced masonry (URM); Strengthening; Welded wire mesh (WWM); Reinforced masonry; Shake table
testing.

Introduction
Almost one-third of Indian houses are made up of unreinforced
masonry (URM), and their performance in the past earthquakes
(Bihar in 1988; Uttarkashi in 1991; Killari in 1993; Jabalpur in
1997; Chamoli in 1999; Bhuj in 2001; Sumatra in 2004; Kashmir
in 2005; Sikkim in 2006 and 2011) has created a necessity to review
the capability of existing structures for future earthquakes, and to
find a suitable strengthening technique to strengthen newly con-
structed masonry structures or to retrofit existing structures. Owing
to low cost and less skilled labor, masonry buildings are still in
use. There are number of studies carried out on strengthening/
retrofitting of masonry structures, both in India and abroad.
Strengthening using fiber-reinforced polymer (FRP), poly-
meric meshes, textile reinforcement, and reinforcing steel wires
(D’Ambrisi et al. 2013; Ashraf et al. 2012; Papanicolaou et al.
2011; ElGawady et al. 2006; Tan and Patoary 2004; Mosallam
2007; Galal and Sasanian 2010; Papanicolaou et al. 2008;
Saleem et al. 2016; Shermi and Dubey 2017) have been reported
by many researchers.
Small-scale and full-scale brick masonry and stone masonry
structures have been tested on a shake table testing facility in
the past, including both reinforced masonry (RM) and unreinforced
masonry (Koutromanos et al. 2013; Lourenço et al. 2013; Saleem
et al. 2016; Stavridis et al. 2016; Candeias et al. 2017). These tests
were conducted to evaluate the performance of masonry to study
the influence of opening, wall aspect ratio, different types of dia-
phragm, reinforcement, and so forth. Nevertheless, no shake table
test has been carried out to evaluate the seismic behavior of
1
Research Scholar, Dept. of Earthquake Engineering, Indian Institute
of Technology Roorkee, Roorkee 247667, India (corresponding author).
E-mail: aqua.shermi3@gmail.com
2
Assistant Professor, Dept. of Earthquake Engineering, Indian Institute
of Technology Roorkee, Roorkee 247667, India. E-mail: dubeyfeq@iitr.ac.in
Note. This manuscript was submitted on March 15, 2017; approved on
July 12, 2017; published online on November 22, 2017. Discussion period
open until April 22, 2018; separate discussions must be submitted for
individual papers. This paper is part of the Journal of Performance of
Constructed Facilities, © ASCE, ISSN 0887-3828.
masonry externally strengthened with welded wire mesh and coarse
sand mortar.
In this study, an attempt has been made to compare the behavior
of conventional unreinforced masonry and masonry strengthened
with WWM and coarse sand mortar by subjecting both the models
to a series of ground motions separately. The strengthened model
was subjected to more intense ground motion than the unstrength-
ened model.
Experimental Program
Unreinforced Masonry Model
A half-scale 2.25 × 2.25 m URM model with 2.0 m height was
built on a steel platform. This was later bolted to the shake table
before testing. Figs. 1 and 2 show the plan and section of the
masonry model. Bricks of size 115 × 57 × 35 mm were used for
making the model (BIS 2000a). A mortar mix of 1∶4 (cement:sand)
was used for constructing the models. The compressive strength of
brick and brick masonry used was 10 and 3.95 MPa, respectively.
Both the models were of a single room size, constructed as per the
common construction practice followed in the state of Uttarakhand,
India. Normally, in case of load-bearing structures, one full-scale
brick wall having thickness 230 mm is constructed. For half-scale
models to be tested on the shake table facility, thickness of the wall
is kept as 115 mm. The same mason was used for constructing the
half-scale models throughout, so the quality of masonry did not
vary because of workmanship. Companion mortar cubes were cast
every time a mortar mix was prepared, to check the quality of
mortar. The north and south wall had a standard door opening
of size 1.50 × 0.75 m. The east wall had a window opening of size
0.75 × 0.75 m, whereas the west wall was a solid wall without any
opening.
The platform for construction of models was made up of flat
steel plate. For providing grip between the foundation and the
model, shear keys were provided with the help of angle iron
(25 × 25 × 3 mm) at the outer and inner periphery. Further, a steel
strip (25 mm wide and 3 mm thick) was welded at the center
of the outer and inner periphery on the plate. A 75-mm-thick

© ASCE 04017121-1 J. Perform. Constr. Facil.

J. Perform. Constr. Facil., 2018, 32(1): 04017121

Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Plan

foundation was laid with 1∶2∶4 concrete mix (BIS 2000b), repre-
senting the fixed base of the model. This arrangement was done (a)
for holding the foundation of the model from shearing/sliding
from the platform during testing. This was equivalent to a rigid
foundation constructed in actual practice. In the absence of the
shear keys, the model may slip or slide because of the flat steel
platform. The shear keys were welded to the platform. The sur-
face of the foundation was roughened with steel wire brush to
have better grip between the foundation and brick work. Before
laying the bricks, a layer of thick grout was applied above the
foundation surface, and then the brick work started above the
75-mm-thick foundation. The sequence of construction was sim-
ilar to that of the conventional construction procedures followed
in the northern part of India. Cut lintel beams were provided
above the door and window openings. The roof slab was con-
structed in the end. The construction procedure of the unrein-
forced masonry model can be seen in Figs. 3–6. The model
was cured for 28 days from the date of casting of slab.

Reinforced Masonry Model

The strengthened/reinforced model had the same dimensions as
that of the unreinforced masonry model. The detailed section of
the retrofitted model is shown in Fig. 7. For constructing a rein-
forced masonry model, a 3-mm-thick steel plate was welded with
the angle iron (shear keys) on all the four corners and at the two (b)
edges of each opening (Fig. 8) at the base of the foundation. The
strengthening of the reinforced masonry model was carried out over Fig. 2. (a) Window-side section URM model; (b) door section URM
an URM model with the provision of the welded wire mesh to be model
anchored with the shear keys through steel plates. This simulates
strengthening of an existing masonry building where the welded
wire mesh is anchored in foundation.
The retrofitting/strengthening work was carried out as per
Indian Standard (IS) code of practice IS:13935-2009 (BIS 2009)
after the complete construction and curing of the conventional
model. The model was cleaned with the help of a steel wire brush
to remove dirt, if any. Then a layer of cement slurry was applied
with a paint brush to provide a better bond between the mortar
and the masonry surface. A 10-mm-thick layer of 1∶3 cement and
coarse sand mortar was applied above the cement slurry surface
to level the uneven masonry surface, as well as to provide a better
grip for the WWM and second layer of mortar. The mortar sur-
face was roughened with the help of a steel wire brush for better
grip. The WWM was placed and anchored with the help of a
Fig. 3. Laying of bricks above foundation
4-mm-diameter mild steel wire. The wires were bent over the

© ASCE 04017121-2 J. Perform. Constr. Facil.

J. Perform. Constr. Facil., 2018, 32(1): 04017121

Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.
Fig. 4. Unreinforced masonry model up to roof level
Fig. 5. Front-facing URM model
WWM on the two sides in opposite directions, as shown in Fig. 8.
These 4-mm wires transfer the shear at the WWM masonry inter-
face through dowel action. A layer of cement slurry was spread
above the WWM and a 10-mm-thick 1∶3 cement:sand mortar was
applied above the WWM and the surfaces were leveled. The
strengthening process can be seen in Figs. 8 and 9. The strength-
ened model was cured under normal site condition for another
28 days.
Free Vibration Test
Free vibration test was carried out to determine the frequency of
the models. For acquiring the free vibration data, two Ranger
Seismometer (SS-1) (Kinemetrics, Pasadena, California) were
placed on the rooftop of the models (Fig. 10)—one in the longi-
tudinal direction (E-W) and other in the transverse direction
© ASCE 04017121-3
J. Perform. Constr. Facil., 2018, 32(1): 04017121
Fig. 6. Fully completed URM model
(N-S). The models were excited by giving mild impact using a
wooden hammer at the base of the models. The data were recorded
with the help of a 12-channel data acquisition system. The free
vibration data of URM and RM models is plotted in Figs. 11
and 12, respectively. The natural time period of the models were
computed from the recorded data. The time period of URM and
RM models was obtained as 0.056 and 0.050 s respectively.
Shake Table Testing
The Department of Earthquake Engineering, Indian Institute
of Technology, Roorkee, has a biaxial shake table facility with
a 20-ton payload capacity with a rigid steel platform of size
3.50 × 3.50 m. This shake table is capable of reproducing specified
real earthquake accelerogram or simulated synthetic accelerogram
compatible with a specified design spectrum. The masonry models
were tested on the shake table for the horizontal component of the
ground motion compatible to IS code of practice IS:1893 (Part 1)-
2002 (BIS 2002) spectra for seismic zone V for hard soil.
For measuring the ground motion transferred to the rigid
platform of the shake table, a triaxial accelerometer (manufac-
tured by Kinemetrics, Pasadena, California, it can record 0.25
to 0.4 g) was fixed with the platform. The models were also
instrumented with triaxial accelerometers at their rooftop for
measuring the acceleration at roof level. In addition to an accel-
erometer on the rooftop, the URM was also instrumented with
LVDT fixed diagonally on the left pier of the shear wall in the
E-W direction, as shown in Fig. 10(b), for measuring the local
displacement and crack width during testing. The LVDT (manu-
factured by Kaptl Instruments, Roorkee, India) used has a
sensitivity of 0.01 mm.
The URM model was subjected to two different horizontal
ground motions similar to that of artificial time history record com-
patible to IS code seismic zone V hard soil spectra (Fig. 13). The
RM model was subjected to four different horizontal ground
motion records. The loading sequence is illustrated in Table 1.
J. Perform. Constr. Facil.
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.
(a)
(b)
Fig. 7. (a) Retrofitted model window-side section; (b) retrofitted model
door-side section
Shake Table Test Results
Behavior of the URM Model
Minor cracks were observed after the first shake to the URM
model, which can be seen in Fig. 14. It was observed that the given
ground motion was not intense enough to create severe damage to
the model. The maximum roof acceleration observed at the top
in the direction of the applied motion (x-direction) was 0.61 g. Plot
of the acceleration time history at the roof in x-direction can be seen
in Fig. 15.
The second ground motion applied to the URM model was
equivalent maximum considered earthquake (MCE). It was ob-
served that the minor cracks formed in the first shake started to
widen further. Severe cracks in the form of shifting/displacements
in the walls at different locations in the URM model (Fig. 16) were
© ASCE 04017121-4
J. Perform. Constr. Facil., 2018, 32(1): 04017121
Fig. 8. Strengthening of model with welded wire mesh
Fig. 9. Reinforced masonry model after completion
also observed. In the north-facing wall, a shift of 7 cm was ob-
served at the lintel level [Fig. 16(a)]. Further, this wall was also
displaced by 4.8 cm [Fig. 16(a)] at the right top corner and by
8.1 cm [Fig. 16(a)] a shift in the first brick layer just below the
roof level. In the west-facing wall, cracks were observed at the cor-
ners of the window opening [Fig. 16(b)]. A shift of 3.5 cm was
observed in the brick layer below the roof at the right top
corner of the wall [Fig. 16(b)]. In the south-facing wall with a door
opening, displacement of the wall of approximately 10 cm was
observed at the left and right lintel corners, with a 5.0-cm-wide
crack. In the east-facing wall (solid wall), a 7-cm shift [Fig. 16(c)]
J. Perform. Constr. Facil.
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.

Fig. 10. (a) Free vibration setup for URM model; (b) free vibration setup for RM model; (c) instrumentation in URM model

was observed at the fourth brick layer below roof. The maximum
roof acceleration observed at the top in the direction of the ap-
plied motion (x-direction) was 0.91 g. The plot of the horizontal com-
ponent of acceleration time history at the roof level in URM model is
shown in Fig. 17 in the direction of the applied motion (x-direction).

Behavior of the RM Model

During the first shake, the maximum roof acceleration observed
at the rooftop in x-direction was 0.60 g. From the recorded accel-
eration data, the velocity and displacements at the rooftop were
computed with the help of SeismoSignal software. The maximum
velocity observed at the roof level in x-directions was 18.61 cm=s.
Fig. 11. Free vibration data of URM model
The maximum roof displacement in x-direction was 9.11 cm. The
horizontal component of acceleration time history at the rooftop
(x-direction) is plotted and can be seen in Fig. 18.
During the second shake, the maximum roof acceleration ob-
served at the top in x was 0.77 g. The maximum roof velocity and
displacement computed at the rooftop level in x-direction were
26.89 cm=s and 9.42 cm, respectively. Fig. 19 shows the acceler-
ation time history recorded at the rooftop in x-direction.
During the third shake, the maximum roof acceleration observed
at the rooftop in x-direction was 0.79 g. The maximum roof ve-
locity observed in x-direction was 17.20 cm=s and maximum
roof displacement in x-direction was 17 cm. Fig. 20 shows the
plot of acceleration time history at the rooftop in x-direction.
During the fourth shake, the maximum roof acceleration ob-
served at the rooftop in x-direction was 0.8 g. The maximum roof
velocity observed at the roof level in x-direction was 26.05 cm=s.
Fig. 12. Free vibration data of RM model
The maximum roof displacement in x-direction was 10.19 cm.

© ASCE 04017121-5 J. Perform. Constr. Facil.

J. Perform. Constr. Facil., 2018, 32(1): 04017121

 Fig. 13. (a) Input ground motion—DBE; (b) input ground motion—MCE
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.

Table 1. Input Ground Motion (Horizontal Component Only) The plotted graphs of acceleration time history at the rooftop in
x-direction can be seen in Fig. 21.
Serial number Unreinforced model Reinforced model
1 DBE DBE
2 MCE MCE Structural Failure Mechanism
3 — 3 times DBE The conventional building model considered in the present study
4 — 4 times DBE
may be treated as a typical box system in which the walls parallel

Fig. 14. (a) Failure pattern in east wall (URM model) after first shake; (b) failure pattern in west wall (URM model) after first shake; (c) failure pattern
in north wall (URM model) after first shake; (d) failure pattern in south wall (URM model) after first shake

© ASCE 04017121-6 J. Perform. Constr. Facil.

J. Perform. Constr. Facil., 2018, 32(1): 04017121


Fig. 15. Acceleration at roof level in x-direction in URM model—DBE
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.
to loading direction are called in-plane walls or shear walls,
whereas the walls perpendicular to the loading direction are termed
as out-of-plane walls or cross walls. The in-plane walls develop a
shear-resisting mechanism, whereas the out-of-plane walls develop
a flexure one. The in-plane walls provide lateral resistance by
means of membrane action. They resist inertia force owing to their
own mass as well as that of the roof, and they also resist the forces
transmitted by out-of-plane walls during shaking. Bending moment
induced in the in-plane walls during shaking is maximum at its
Fig. 16. (a) Failure pattern in north wall (URM model) after second shake; (b) failure pattern in west wall (URM model) after second shake; (c) failure
pattern in east wall (URM model) after second shake
© ASCE
J. Perform. Constr. Facil., 2018, 32(1): 04017121
extreme ends producing vertical cracks at the corners. The diagonal
cracks in the in-plane walls leading to its shear failure are due to the
principal tensile stresses developed in the walls because of vertical
and lateral loads. Owing to the window and door opening, the un-
supported length of the shear wall is reduced, accordingly its shear
strength is reduced. The shear forces become critical in the in-plane
walls with openings and the failure occurs in the form of diagonal
cracks.
In the URM model, cracks initiated at the plinth level and propa-
gated toward the top of the model. During the second shake, the
URM model was severely damaged, causing widespread gaps in
walls, and the model was almost at the verge of collapse. In case
of RM model (Fig. 22), no cracks were seen because of the pres-
ence of sill band, lintel band, and vertical band (in form of WWM)
due to the additional tensile and bending strength developed in
the model. Thus, the failure of conventional brick masonry can
be greatly minimized by incorporating the adequate earthquake-
resistant features.
Summary of Observation
Unreinforced masonry and RM models were constructed on a
steel base plate, and the tests were performed on the shake table.
04017121-7 J. Perform. Constr. Facil.
 Fig. 17. Acceleration at roof level in x-direction in URM
model—MCE
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.
Fig. 18. Acceleration at roof level in x-direction in RM model—DBE
Fig. 19. Acceleration at roof level in x-direction in RM model—MCE
Fig. 20. Acceleration at roof level in x-direction in RM model—
3 times DBE
During the time of testing, behavior and response of both the mod-
els subjected to dynamic loading were examined. The URM model
was constructed as per conventional procedure, whereas the RM
model was constructed as per the IS code provisions for seismic
© ASCE 04017121-8
J. Perform. Constr. Facil., 2018, 32(1): 04017121
Fig. 21. Acceleration at roof level in x-direction in RM model—
4 times DBE
Fig. 22. Undamaged RM mode after testing
zone V. The earthquake-resistant features included reinforcement
in the form of WWM at sill level, lintel level, and at all the four
corners and near the openings.
Total of two ground motions were applied to the URM model.
Third shaking was not given so as to avoid collapse of the model,
which could have damaged the components/parts of the shake
table. In the RM model, total of four ground motions were applied
with increasing intensity. Triaxial accelerometers were used to re-
cord the base motion of the table and the roof motion of the models.
Minor cracks were observed in the URM model after the first
shake (Fig. 14) and the cracks widened further during the second
shake, resulting in severe damage to the model leading to the verge
of its collapse. The inertia force transferred from the roof to the
shear walls on the northern and southern side led to the develop-
ment of cracks originating near the openings. Although the in-plane
shear walls offer greater resistance because of their larger depths,
unreinforced masonry shear walls developed horizontal cracks at
their base and around the window and door openings.
The RM model was subjected to intense ground motions than
the URM model. The earthquake-resistant features provided in the
strengthened model in the form of sill band, lintel band, and vertical
corner band in all the four walls contributed in transferring the
out-of-plane inertia force to the shear walls. The walls have been
effectively tied together by placing vertical reinforcements to avoid
the separation of vertical joints at the corners during shaking. North
and south shear walls can resist lateral forces owing to their own
mass and those transmitted to it from the roof. Further, the roof
was properly anchored to the walls with the help of vertical rein-
forcement at all the four corners and near the openings starting from
the foundation level, for exhibiting diaphragm action. All these
J. Perform. Constr. Facil.
 earthquake-resistant features incorporated in the strengthened
model have not created any damage in the model even after the
fourth shaking (higher than MCE).
Conclusions
Based on the foregoing experimental studies, following conclu-
sions are drawn.
1. The URM model designed for gravity loads can resist only mild
earthquakes. The URM model subjected to earthquake excita-
tion corresponding to design basis earthquake (DBE), exhibited
minor cracks in the walls because the ground motion was not
Downloaded from ascelibrary.org by The University of Manchester on 03/12/18. Copyright ASCE. For personal use only; all rights reserved.
intense in comparison, and the model was able to dissipate
the energy.
2. The URM model, when subjected to earthquake excitation
corresponding to MCE, developed major cracks and severe
damage, which was in close agreement with the behavior of
masonry in a real earthquake situation.
3. The RM model performed well both during DBE and MCE
conditions which was clearly evident during the testing. The
strengthening of the conventional model by incorporating
earthquake-resistant elements externally made the model stiff
and ductile enough to dissipate the energy without causing any
damage to it.
4. The RM model did not show any hairline cracks even though
it was subjected to more intense time history acceleration than
the URM model. There was an almost 125% increase in perfor-
mance in the RM model when compared with the URM model.
The RM model behaved ductile compared with the URM model.
5. The proposed techniques have reaffirmed the effectiveness of
incorporating the earthquake-resistant provisions given in the
IS codes of practice for strengthening the earthquake-deficient
existing masonry structures.
References
Ashraf, M., Khan, A. N., Naseer, A., Ali, Q., and Alam, B. (2012).
“Seismic behavior of unreinforced and confined brick masonry walls
before and after ferrocement overlay retrofitting.” Int. J. Archit.
Heritage, 6(6), 665–688.
BIS (Bureau of Indian Standards). (2000a). “Code of practice for prepara-
tion and use of masonry mortars.” IS 2250, New Delhi, India.
BIS (Bureau of Indian Standards). (2000b). “Plain and reinforced concrete
code of practice.” IS 456, New Delhi, India.
BIS (Bureau of Indian Standards). (2002). “Criteria for earthquake resistant
design of structures.” IS 1893, New Delhi, India.
© ASCE 04017121-9
J. Perform. Constr. Facil., 2018, 32(1): 04017121
BIS (Bureau of Indian Standards). (2009). “Indian standard seismic
evaluation, repair and strengthening of masonry building—Guidelines.”
IS 13935, New Delhi, India.
Candeias, P. X., Campos Costa, A., Mendes, N., Costa, A. A., and
Lourenço, P. B. (2017). “Experimental assessment of the out-of-plane
performance of masonry buildings through shaking table tests.” Int. J.
Archit. Heritage, 11(1), 31–58.
D’Ambrisi, A., Feo, L., and Focacci, F. (2013). “Experimental and ana-
lytical investigation on bond between Carbon-FRCM materials and
masonry.” Compos. Part B, 46, 15–20.
ElGawady, M. A., Lestuzzi, P., and Badoux, M. (2006). “Aseismic retro-
fitting of unreinforced masonry walls using FRP.” Compos. Part B,
37(2), 148–162.
Galal, K., and Sasanian, N. (2010). “Out-of-plane flexural performance
of GFRP-reinforced masonry walls.” J. Compos. Constr., 10.1061
/(ASCE)CC.1943-5614.0000061, 162–174.
Koutromanos, I., Kyriakides, M., Stavridis, A., Billington, S., and Shing,
P. B. (2013). “Shake-table tests of a 3-story masonry-infilled RC frame
retrofitted with composite materials.” J. Struct. Eng., 10.1061/(ASCE)
ST.1943-541X.0000689, 1340–1351.
Lourenço, P. B., Avila, L., Vasconcelos, G., Alves, J. P. P., Mendes, N.,
and Costa, A. C. (2013). “Experimental investigation on the seismic
performance of masonry buildings using shaking table testing.” Bull.
Earthquake Eng., 11(4), 1157–1190.
Mosallam, A. S. (2007). “Out-of-plane flexural behavior of unreinforced
red brick walls strengthened with FRP composites.” Compos. Part B,
38(5), 559–574.
Papanicolaou, C. G., Triantafillou, T. C., Papathanasiou, M., and Karlos, K.
(2008). “Textile reinforced mortar (TRM) versus FRP as strengthening
material of URM walls: Out-of-plane cyclic loading.” Mater. Struct.,
41(1), 143–157.
Papanicolaou, C., Triantafillou, T., and Lekka, M. (2011). “Externally
bonded grids as strengthening and seismic retrofitting materials of
masonry panels.” Constr. Build. Mater., 25(2), 504–514.
Saleem, M. U., Numada, M., Amin, M. N., and Meguro, K. (2016). “Shake
table tests on FRP retrofitted masonry building models.” J. Compos.
Constr., 10.1061/(ASCE)CC.1943-5614.0000684, 04016031.
SeismoSignal [Computer software]. Seismosoft, Pavia, Italy.
Shermi, C., and Dubey, R. N. (2017). “Study on out-of-plane behaviour of
unreinforced masonry strengthened with welded wire mesh and mortar.”
Constr. Build. Mater., 143, 104–120.
Stavridis, A., Ahmadi, F., Mavros, M., Shing, P. B., Klingner, R. E., and
McLean, D. (2016). “Shake-table tests of a full-scale three-story rein-
forced masonry shear wall structure.” J. Struct. Eng., 10.1061/(ASCE)
ST.1943-541X.0001527, 04016074.
Tan, K. H., and Patoary, M. K. H. (2004). “Strengthening of masonry walls
against out-of-plane loads using fiber-reinforced polymer reinforce-
ment.” J. Compos. Constr., 10.1061/(ASCE)1090-0268(2004)8:1(79),
79–87.
J. Perform. Constr. Facil.