This article was downloaded by: [York University Libraries]

On: 23 September 2014, At: 14:28

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK

Architectural Science Review

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tasr20

Evaluation of the Structural Safety of Historical

Masonry Buildings
Ali hsan nay

Department of Architecture , Middle East Technical University , TR-06531, Ankara,

Turkey Phone: +90 312 2102251 Fax: +90 312 2102251 E-mail:
Published online: 09 Jun 2011.

To cite this article: Ali hsan nay (2007) Evaluation of the Structural Safety of Historical Masonry Buildings, Architectural
Science Review, 50:1, 26-30
To link to this article: http://dx.doi.org/10.3763/asre.2007.5004

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

www.earthscan.co.uk/journals/asre

doi:10.3763/asre.2007.5004

Architectural Science Review

Volume 50.1, pp 26-30

Evaluation of the Structural Safety of Historical

Masonry Buildings
.

..

Downloaded by [York University Libraries] at 14:28 23 September 2014

Ali Ihsan Unay

Department of Architecture, Middle East Technical University, TR-06531 Ankara, Turkey
Tel: +90 312 2102251; Fax: +90 312 2101249; E-mail: unayacademic@yahoo.com
Received 22 May 2006; accepted 13 September 2006

Abstract: Structural stability problems due to the formation of cracks and local collapses are the governing criteria for the structural safety
of historical masonry buildings. The character of these cracks can be studied by numerical analysis. Analytical studies of the structural
preservation of historical structures have two aspects. Firstly, structural demand should be determined by a decisive structural analysis.
Second, the strength capacity of the structural elements must be evaluated. A comparison of these two concluding stages results in the
determination of the ratio of safety. In this paper, several steps of analytical study are proposed to evaluate the safety level of historical
structures. Structural demand of any given structure can be determined by using nite element analysis, which can be used to single
out the vulnerable parts of the structure. Then, an analytical method can be derived to calculate the strength capacity of these structural
elements. In this method, the interaction relationship between bending moment and the axial force is generated by using the material
properties of masonry. To illustrate the methods, a case study is performed on the Yivli Minaret in Antalya, Turkey, originally part of a
Byzantine church converted into a mosque in 1230, and one of the oldest examples of multicupola construction in Anatolia.
Keywords: Historical structures, Safety, Masonry construction, Finite element analysis

Introduction
The main principle of structural conservation is to protect structures against environmental and man-caused hazards. Repair and/or
restoration may be necessary for damaged structures. Historical
structures are strengthened mostly because of the deterioration in
their structural material due to the eects of time. Most of the
historical structures are, unfortunately, in poor condition. Natural
disasters, foundation settlements and res have taken their toll.
Moreover, centuries of neglect have eroded away many treasures.
An important percentage of historical structures, at present,
have serious cracks and deterioration problems that threaten their
structural integrity. These are often accompanied by partial or
almost full collapse. Primitive and unskilled attempts have been
made to seal cracks and to repair roofs and walls in particular. But,
the missing and the most important point here is that any attempt
that is performed before a full conception of the reasons for the
formation of these faults could be inappropriate, even dangerous.
Both the original and the presently defective load transfer mechanisms of the structure must be well studied before deciding on the
rehabilitation, restoration or strengthening techniques. Otherwise,
it might not be so pretentious to claim that any preservation eort
might even unduly harm the historical structure, let alone help
it (Raman, 2004).

Under these circumstances, it could be said that, before any

attempts related to conservation or preservation of historical buildings, the initial step should be to develop methods of structural
analysis for these structures in order to understand their load
transfer mechanisms. Analytical models that are as close as possible
to actual dimensions and structural conguration of the structure
in question should be developed. On such a model, various loads
(natural and articially induced) can be applied and their actions
can be analysed. Load carrying mechanism can be understood
through such analyses (Kuhlmann, 2004). The architect and
engineer would then be in a position to attempt an appropriate
method of repair. They would be aware of the forces that are in
equilibrium and would know the susceptible parts of the structure.
Thus, they would be able to assess the eects of the repair methods,
which will change, modify and improve the present forces that are
in some kind of equilibrium (Loureno, 2002).

Structural Safety of Masonry and

Evaluation of the Ratio of Safety
Safety of a structure is generally dened as the reserve resistance to failure. Evaluation of structural safety is frequently based
on the requirement that a margin of safety is needed to cover

Ali Ihsan Unay

The Structural Safety of Historical Masonry Buildings

27

characteristic under special conditions. However, because of the

inherent material properties, cross-sectional properties and overall
structural form, historical masonry structural elements are subject
to axial compressive forces and bending moments (Ochsendorf,
Hernando Huerta, 2004). If the ultimate capacity of structural
elements under axial force and bending moment are known, the
generated N-M interaction diagram can be meaningful, if the element is assumed to have linearly elastic material properties.

Downloaded by [York University Libraries] at 14:28 23 September 2014

Figure 1: Evaluation of the safety factor using an N-M

interaction diagram.
uncertainties in loading, materials, and methods of analysis and
methods of construction. Structural safety is strongly dependent
on intuition and judgment based on experience, so that safety
decisions cannot be made exclusively with analytical models of
probability and mechanics.
The measure of safety is called the factor of safety. The factor
of safety can be dened as, the strength capacity of the structural
element, divided by the force eect, which would be subjected
to the structural element. It is not possible to predict the exact
maximum value of loads that a structure will experience during
its lifetime. A reasonable prediction of the maximum expected
value based on past records can be considered. Loads may be
dened as random variables. The resistance of a structure is
also a random variable. Variable strengths can exist in dierent
stone or brick elements even though they have the same sort
of material grade or class (Meskouris, Butenweg, Mistler &
Kuhlmann, 2004).
Structural safety and factor of safety can be expressed by
considering probability of failure of a structure. In determining
the safety factor of historical masonry structures, various criteria
should be considered. Consequently, it is not so easy to reach an
exact solution by conventional engineering analysis. In order to
estimate the safety of a historical masonry structure, behaviour of
the structure must be determined well. Principally, geometrical
form of structure, materials of construction, loading and support
conditions must be studied with as much precision as possible
(Giordano, Mele & De Luca, 2001).
If the failure of an individual element causes overall collapse
of the structure, it becomes important to detect at what time and
for what reason this element fails. Masonry structural elements
frequently have nonlinear material properties and the uniformity
of material characteristics is uncertain along the entire length of
the element. For such cases, the exact solution for the evaluation
of safety is to determine the ultimate strength of the element.
Therefore, if stress-strain relationship is known, the ultimate
strength of any masonry cross-section can be determined (Corradi,
Borri & Vignolib, 2003).
Axial force (N), bending moment (M), shear force (V) and
torsional moment (T) are the main internal forces that inuence
the masonry elements. Torsion and shear forces may be more

The N-M interaction diagram shown in Figure 1 illustrates

the safety range of a masonry element. As the result of structural
analysis, axial force NA and bending moment MA is determined at
point A on the interaction diagram. Since the structure is analyzed
by assuming linear elastic material characteristics, gradually increasing loads lead to linear increase in the internal forces. Therefore, a
line connecting the origin and the point A intersect the interaction
diagram at point B. At point B, the element is assumed to have
reached its load carrying capacity. Consequently, the safety range can
be dened on the curve. Indeed, in addition to the mathematical
meaning of this ratio, graphical signicance of this ratio is more
meaningful for the concept of safety evaluation.
This approach facilitates the analysis of historical masonry
structures for safety investigation and exploration of structural load
carrying capacity under various disturbances. Structural safety of
historical masonry buildings is dependent on several factors and
thus, interpretation of these factors is very dicult. Therefore, a
systematic study is needed for the evaluation of the safety level.

Determining the Strength Capacity of

Masonry
This study outlines an analytical method, which is developed
to evaluate the strength capacity of any given masonry element
section under combined axial force and bending moment. The
best way to dene the strength capacity of masonry is to plot axial
load (N) bending moment (M) interaction diagram for the element section. In determining the strength capacity of masonry
structural elements analytically, some basic assumptions should be
considered. In this analytical study, both moment and axial force
load carrying capacities are determined simultaneously. Basic assumptions in the determination of combined exural strength of
masonry structural elements are as follows:
Plane sections before bending remain plane after bending.
There is a perfect bond between distinctive materials such as
masonry unit and mortar.
The strain in the masonry units is equal to the strain in the
mortar at the same level.
The stress-strain curve of the material is known.
The stress in the masonry unit and mortar can be computed
from strain using the stress-strain curves for masonry materials.
Tensile strength of material is also to be considered in section
analysis.

28

Architectural Science Review

The strain distribution in the section is uniform under pure

axial force.
The strain distribution in the section is linear under exure.

Downloaded by [York University Libraries] at 14:28 23 September 2014

Masonry is assumed to fail when the compressive strain reaces

0.003 at ultimate state.

Volume 50, Number 1, March 2007

There are dierent material characteristics of masonry structural elements and dierent stress-strain relationship of distinct
materials. This should be considered in the determination of
load carrying capacity of elements due to exure. In order to
determine the ultimate strength of masonry structural elements,
stress-strain relationship of materials must be known. This can
be achieved by testing material samples of the structural element
(Corradi et al., 2003).
The use of stress-strain (-) relation, equilibrium and compatibility conditions in the section
are a quick and simple way to evaluate the load
carrying capacity of masonry structural elements.
In this study, a method which is based on the estimation of the load carrying capacity of the section
for any combination of axial force (N) and bending
moment (M) in the structural element, namely NM interaction diagram of the section is proposed
(Unay, 2002).

Figure 2: A masonry element subjected to axial force.

In historical masonry structures, almost all

of the structural elements are subjected to both
axial load and bending moment. There is almost
no structural element, which is under axial force
without bending moment. Therefore, combined
eect of axial force and bending moment must be
considered in determining the load carrying capacities of elements.
The strength of the masonry element subjected
to pure axial force should be considered as shown
in Figure 2. On the other hand, strength of a general masonry element subjected to exure can be
determined as shown in Figure 3. At any section,
the stress, when integrated over the section must
add up to the required force eects M and N as
shown in Figure 4.

Figure 3: Masonry element subjected to bending.

Figure 4: Equilibrium of internal forces in a masonry element subjected to axial

force and bending moment.

In order to determine the strength capacity of

a masonry section, an axial force (N) / bending
moment (M) interaction diagram is plotted for
the section. As shown in Figure 5, the element
cross-section is dened by a nite number of layers parallel to main bending axis. The number
of layers that is going to be used is related to the
dimensions and shape of the structural element.
Furthermore, while modelling elements, distinct
kinds of materials can be used for each dierent
layer. The voids in the cross-section may also be
considered. Material characteristics of the crosssection are dened in terms of a stress-strain (-)
curve. Since the axial force (N) and bending moment (M) carrying capacity of masonry structural
elements with arbitrary geometry can be analyzed
easily, safety factors of these structures can also be
determined in an approximate manner.
In association with an N-M interaction diagram,
the load carrying capacity of all kind of masonry

Ali Ihsan Unay

The Structural Safety of Historical Masonry Buildings

29

Downloaded by [York University Libraries] at 14:28 23 September 2014

Figure 5: Denition of element cross-section by a nite number of

layers.
element cross-sections can be evaluated by satisfying the equilibrium conditions in the cross-section. Masonry structures have
unusual cross-sectional shapes because of their architectural form
and construction techniques. By dividing the overall section into
small rectangular or triangular parts, all kind of irregular element
cross-sections and dierent material characteristics in cross-section
of certain masonry elements could be designated by nite number
of regions. Dierential areas should be taken into account to satisfy
the equilibrium and compatibility condition in the section.
Employing adequate number of sub-areas according to the shape
and material characteristics of the section, the force and equilibrium
condition can be satised. As shown in Figure 4, equilibrium
equation can be written as according to location of neutral axis
N = dA1 1 + dA2 2 + = dAi i

(1)

In this equation, can be evaluated in two steps. Initially the

strain (i) corresponding to dierential area (dAi) at this level is
determined from strain distribution in the section, and then, is
determined by using the - curve of the identical material.
Consequently, equilibrium state of computed internal forces
is ensured in the cross-section from Eq. (1), and then internal
moment equation can be written as
M=Fi yi

(2)

Applying this procedure, axial force N and bending moment M

values in the section can be determined. Moreover, tensile and
compressive crushing strain crushing should not be exceed in
the strain variation in the section at uppermost and lower most
bre of the section. Exceeding the crushing strain crushing at
the section means failure of the masonry element. Calculated
axial force N and bending moment M can be considered as the
maximum load carrying capacity (Unay, 2002).

Application of the Method to a Case Study

In order to demonstrate the versatility of the method, the most
unusual cross-sectioned example is selected (Figure 6). A case
study was performed on the Yivli Minaret in Antalya, Turkey.

Figure 6: Yivli Minaret with its unusual crosssection. Photograph by the author.
The Yivli Minaret Mosque was originally a Byzantine church
and was converted into a mosque in 1230 by Seljuks. Six semispherical cupolas categorize it as one of the oldest examples of
multicupola construction in Anatolia. Besides, the minaret is the
oldest structure of an informal religious complex located in the
historic castle of Antalya. The 38 m high brick masonry minaret
is located on a stone base with its 3 m diameter. It was restored in
1953 and was structurally stabilized in 1973 by state authorities
(Akurgal, 1980).
The minaret is investigated analytically by nite element analysis.
The numerical model reecting the whole structure as presented
in Figure 7 is adopted with an idealization approach of assembling
2400 eight-noded nite elements. The following properties have
been assumed in the analyses: Modulus of elasticity 10 103
N/mm2 and Poisson ration of 0.2.
The structure was analysed under gravity loads and a maximum
credible earthquake of base-rock acceleration of 0.4 g by the
Response Spectrum analysis using SAP2000 Structural Analysis
software (SAP 2000, 2002). The normalized spectrum cure was
obtained from the current Turkish Earthquake Code, considering
medium dense sand gravel, which is classied as soil type C in the
code. The natural periods of the minaret in the rst three modes
were calculated as T1= 0.54 sec., T2=0.53 sec. and T3=0.12 sec.,
where the total mass of the structure is m=794 kN.
Results from the computer analysis show that a remarkable
behaviour is achieved in the minaret. Numerical outputs of the
analysis are interpreted and by considering a single column analogy, they are converted to axial force N and bending moment.
By using the results of the gravity analysis, Ng= 4200 kN and
Mg= 840 kNm, (Ng, Mg), a point is plotted in the interaction
diagram, which shows the equilibrium position under gravity
loads. By considering this point of equilibrium as the origin, a

30

Architectural Science Review

Volume 50, Number 1, March 2007

Downloaded by [York University Libraries] at 14:28 23 September 2014

Figure 8: Determination of the safety level.

and seismic loads can be plotted on the interaction diagram to
determine the safety factor

Figure 7: Finite element analysis of the Yivli Minaret.

second point (Ng+Neq, Mg+Meq) is plotted by using the results
of the earthquake analysis, Neq= 7650 kN and Meq= 7062 kNm.
This second point takes place out of the interaction diagram
pointing out the failure of the structure. The intersection point
is labelled as the failure point. The ratios of the distance between
AD to AB give the safety range under the considered earthquake.
As shown in Figure 8, the minaret is extremely safe under the
gravity loads by a safety factor of SF=11. On the other hand,
it would fail due to an earthquake of 0.4 g of ground acceleration. The intensity of failure earthquake is determined as 0.33 g.
This safety analysis shows that, Yivli Minaret can only resist an
earthquake with 0.33 g of ground acceleration. It has therefore
been found that under an earthquake of base-rock acceleration
of 0.4 g, the minaret is unsafe.

Conclusions
Safety factors may be dened as the ratio of the strength capacity
of the historical structure to the load eects imposed. Therefore,
in order to dene the safety level rationally, a well-performed
structural analysis is needed. Finite element methods provide the
most appropriate means to realize these analyses.
The strength capacity of the historical structure is determined
by developing the axial force N / bending moment M interaction
diagram of the cross-sections that are selected as vulnerable by the
extensive analyses already performed. Using realistic stress-strain
relationships for masonry, the interaction diagram between axial
forces and bending moment can be, and, in the case study, was
determined. The combination of (N, M) due to gravity loads

Repair and rehabilitation process can be done with much greater

condence, once the vulnerable parts of the structure are known
and the factor of safety quantied. The methodology to dene
the inherent safety level of historical structures thus becomes an
invaluable tool for the engineer and architect who attempt to repair
or rehabilitate historical structures.

Acknowledgment
The author thanks very much the referees and editor for their great
contributions to the paper.

References
Akurgal, E. (1980). The Art and Architecture of Turkey. New York: Rizzoli.
Corradi, M., Borri, A., & Vignolib, A. (2003). Experimental study on
the determination of strength of masonry walls. Construction and
Building Materials, 17, 325337.
Giordano, A., Mele, E., & De Luca A. (2001). Modeling of historical
masonry structures: Composition of dierent approaches though a
case study. Engineering Structures, 24, 1057-1069.
Kuhlmann, W. (2004). Historic buildings under earthquake load. IABSE
Reports, 88, 364-365.
Loureno, P. B. (2002). Computations on historic masonry structures.
Progress in Structural Engineering Materials, 4, 301-319.
Meskouris, K., Butenweg, C., Mistler, M., & Kuhlmann, W. (2004).
Seismic behaviour of historic masonry buildings. Proceedings of 7th
National Congress on Mechanics of HSTAM, Chania, Crete, Greece,
24-26 June 2004 (published in CD proceedings).
Ochsendorf, J. A., Hernando, J. A., & Huerta, S. (2004). Collapse of masonry buttresses. Journal of Architectural Engineering, 10(3), 88-97.
Raman, P. G. (2004). Structural masonry and architectural expression.
Construction and Building Materials, 18, 133-139.
SAP 2000. (2002). Version 8, Analysis Reference Manual. Berkeley, Calif:
Computers & Structures Inc.
Unay, A. I. (2002). Earthquake Response of Historical Structures. Ankara:
Middle East Technical University Faculty of Architecture Press.