Engineering Failure Analysis 115 (2020) 104662

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

Seismic vulnerability assessment methodology for historic masonry

T
buildings in the near-field areas
⁎
Marius Mosoarcaa, , Iasmina Onescua, Eugen Onescub, Anthimos Anastasiadisc
a
Research Center for Architecture and Urban Planning, Faculty of Architecture and Urban Planning, Politehnica University of Timisoara, Traian
Lalescu no. 2/A, Timisoara, Romania
b
Faculty of Architecture and Urban Planning, Politehnica University of Timisoara, Traian Lalescu no. 2/A, Timisoara, Romania
c
GeoStatic, Geotechnical-Structural Department, Thessaloniki, Greece

A R T IC LE I N F O ABS TRA CT

Keywords: Timisoara is the biggest city located in the Banat seismic area, the second most important seismic
Failure mechanism zone of Romania. The specificity of the area is represented by shallow earthquakes of crustal
Earthquake type, with small focal depths and a PGA = 0.20 g.
Near-field Because Timisoara was declared the European Capital of Culture 2021, it is mandatory to
Masonry
assess the seismic vulnerability of its most sensitive areas and to design potential losses scenario.
Historical building
This study also provides useful data for the local authorities, helping them develop/improve the
prevention and intervention plans.
The paper focusses on the analysis of two historical urban districts, likely to be among the
most attractive tourist areas. The study aims to assess the seismic vulnerability of the districts in a
quick and simplified way, seeks to identify the exposure of the city, and also to provide losses
statistics for a specific seismic scenario. The empirical seismic vulnerability assessment metho-
dology is based on European studies and is applied for more than 100 historic buildings.
Our assessment analysis aims to adapt the existing seismic assessment methodologies to the
near-field earthquake’s effects. That is why a new formula is proposed to correlate the empirical
vulnerability curves with the actual damage level. The real damage state was observed on similar
historical masonry buildings during the site inspection and after the past earthquakes in nearby
areas. Besides, a new failure mechanism is highlighted based on the investigation of the effects of
previous near-field earthquakes in the Banat seismic region. The new methodology proposes
original vulnerability curves, particularly for the near-field earthquakes specific to Banat seismic
area.

1. Introduction

Hazards have always represented a risk factor, either because of various natural events, such as earthquakes, floods, tsunamis,
extreme winds, volcano eruptions, or because of other factors. While hazards are almost impossible to be predicted or influenced,
how a city/ an area responds to them is significantly influenced by the risk reduction policies. The vulnerability of an urban area
represents a sum of hazard and exposure, the latter being highly dependent on the urban development strategies, the prevention and
intervention plans.

⁎
Corresponding author.
E-mail addresses: [email protected] (M. Mosoarca), [email protected] (I. Onescu), [email protected] (E. Onescu),
[email protected] (A. Anastasiadis).

https://doi.org/10.1016/j.engfailanal.2020.104662
Received 16 October 2019; Received in revised form 1 June 2020; Accepted 2 June 2020
Available online 06 June 2020
1350-6307/ © 2020 Elsevier Ltd. All rights reserved.
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 1. Cultural promenade proposed for Timisoara European Capital of Culture 2021.

Protecting the cities, it is a difficult task, mainly because of the high level of uncertainty specific to hazards. The level of vul-
nerability is dependent on the hazards and of the level of exposure, which can be influenced by human actions and local authority
policies.
Considering vulnerability as the susceptibility level to damage or injury highlights the need to anticipate the risk to build a plan
for risk reduction and the recovery from an impact of a specific natural hazard [1].
Romania presents two major seismic zones, Vrancea and Banat. The latter is the second most important seismic area and is located
near Timisoara. This city ranges among the biggest cities of Romania, after the capital, Bucharest. The main seismic characteristic of
the area is the crustal type earthquake, with small focal depths [2].
The cultural strategy of Timisoara helped it win the competition for the European Capital of Culture 2021. This city has always
focused on promoting the interaction between the local communities and the tourists through various cultural events hosted in its
public spaces.
The historical areas are also the most important ones in Timisoara and are considered to be proof of the city’s history and not just
the main attraction poles [3]. That is why many cultural events are to happen in this old part of the town. There was proposed a
cultural promenade, as presented in Fig. 1 [4], which has to be investigated as far as the risk factors are concerned to ensure the safety
of the tourists and local community.
Because Timisoara is located in Banat seismic region and it does have many buildings that are in the protected historical area,
which will be very visited in 2021, there is highlighted the need to assess their seismic vulnerability. To perform such an assessment
at an urban scale is necessary to propose a hybrid methodology. This proposed methodology follows an existing empirical seismic
vulnerability assessment methodology that is used in entire Europe [5] but adapts it to the particularities of the Banat seismic region
and seismic performance of the historical masonry buildings in Timisoara. This calibration aims to evaluate the vulnerability of a
large historical area by performing simplified numerical analysis on only the most representative buildings. This particular adap-
tation represents the main difference between the proposed methodology and the classical FEM approach [6]. Such an evaluation
could serve a useful tool for the local authorities by providing a prioritization list for the most necessary rehabilitation works. In
comparison with the DEM approaches, the methodology is appropriate for Timisoara because, in Romania, there is no such database
that could provide the necessary information. Moreover, the necessary computational input effort for the DEM approach is much
higher, relying on advanced numerical analysis [7–10]. In conclusion, the proposed methodology represents a simplified alternative
to the existing numerical and hybrid seismic vulnerability assessment methodologies that can be easily applied at a large scale. Still, it
should be used only as a first and quick step of the vulnerability assessment, but to obtain a more precise calibration, sophisticated
numerical analysis [11–13] together with expert’s evaluation should be carried out for the most representative historical masonry
buildings in Timisoara.

2. Case study area

2.1. Timisoara historical evolution

Nowadays, Timisoara ranks as the second most important city of Romania, a pole for economic, educational, and cultural activity.
Its location is in the western part of the country, as shown in Fig. 2 [14].
Even if the first signs of human existence on the land of Timisoara date back from the Neolithic, the early official recognition of
the town is only in the 13th century. The documents mention it as a fortified fortress and one century later as a town [14]. From the
trade point of view, the settlement continued to increase under Hungarian administration until the year 1552 when the city fell under
Ottoman administration. This period also coincides with the first urbanistic approach, which consisted of dividing the town into
central districts and suburbs. This type of administrative organization ceased to exist in 1716, when Timisoara was under the Austro-
Hungarian administration [2], see Fig. 3a [15]. At the beginning of the 18th century, a long process of modernization and sys-
tematization started, and a new fortress in the Vauban architectural style was built (Fig. 3b [15]).

2
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 2. Localization of Timisoara city.

Fig. 3. Timisoara city plan: (a) made in 1716; (b) made in 1808.

The center of the city remained inside the fortress walls and was called the ‘Cetate’ (The Citadel) district. As far as the suburban
areas are concerned, they were confined outside the walls at a strategic defense distance from the fortress [2,16], and were called
‘Iosefin’ and ‘Fabric’ districts. In time, the safe distance between the buildings was occupied by new constructions, made of masonry
and reinforced concrete, which led to a practical unification of the historical districts, see Fig. 4a [17]. Thus, the development of the
historical areas was influenced by the various cultural influences of the following architectural styles: Secession, Art Nouveau, and
Baroque.
It is important to stress out that in the period of the construction of these buildings, there were no official antiseismic design rules.
Therefore, there was no need to anticipate the response of the structures to potential earthquakes. That is why the two major seismic

Fig. 4. Timisoara map: (a) the historical area in the context of the nowadays city; (b) seismic faults in Timisoara.

3
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 5. Correspondence between peak ground acceleration and macroseismic intensity EMS-98 [20].

faults in the western part of the city (Fig. 4b) [2] were not taken into consideration. Our study focusses on the seismic risk of the
historic buildings located in a city placed at the intersection of two seismic faults.

2.2. Seismicity of the area

The Banat seismic area is characterized by crustal earthquakes with small focal depths and magnitudes between Mw = 0.2–5.6
[18]. According to the Romanian Design Code P100-1/2013 [19], the peak ground acceleration for Timisoara is PGA = 0.20 g.
According to Eq. (1) and Fig. 5 [20], there was determined the fact that the most probable macroseismic intensity would be IX EMS-
98 [21].

ln(PGA) = 0.24 × IEMS − 98 − 3.9 (1)

In the last 150 years, there were registered several earthquakes with intensities more than V MSK scale, as presented in Table 1
[22]. The data highlight the potential seismic risk for the area.
The maximum registered magnitude was Mw = 5.6, in the epicenters of Banloc, Voiteni, and Ciacova [23]. Based on the at-
tenuation law described in Eqs. (2)–(4), there were determined the most probable MSK [24] and EMS-98 [25] intensities for the area,
as presented in Table 2 [23].

IEMS − 98 = 1.45 × MW − 2.46 × ln(R) + 8.166, (2)

R= d 2 + hf2 [km] (3)

IMSK = 1.5 × MW − 3.5 × log (d) + 3, (4)

where d represents the epicentral distance and hf represents the focal depth.
Considering the PGA = 0.20 g, Fig. 5 and Eqs. (1)–(4), there are multiple possible scenarios. This paper will consider two different
scenarios that are likely to occur in Timisoara city. The first scenario considers an Mw = 4, d = 5 km, and hf = 5 km and could
happen in case of activation of the existing seismic faults in the western part of the city. In this scenario, the most probable mac-
roseismic intensity is again IX EMS-98. The same macroseismic intensity could also be felt in case of an earthquake with MW = 5,
d = 10 km, and hf = 10 km, in case of activation of seismic faults outside the city. We did not consider magnitudes of MW = 6 since
there are no recordings that indicate a similar magnitude in Timisoara.
Considering the results of Eq. (1) and Table 2, there was discussed further in this paper the macroseismic intensity IX EMS-98 for
the study.

Table 1
Main registered earthquakes in Banat region according to MSK intensity.
Seismic intensity V VI VII VII–VIII

Year 1889 1973 1859 1879

1896 1879 1915
1902 1900 1991
1907 1941
1950 1959

4
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Table 2
Correlation between magnitude and seismic intensity.
Mw d [Km] hf [Km] I [EMS-98] I [MSK]

4 5 5 9 7
10 10 7 6
15 15 6 5
20 20 6 5
25 25 5 4
5 5 5 11 8
10 10 9 7
15 15 8 7
20 20 7 6
25 25 7 6

2.3. Typical failure mechanisms for the near-field earthquakes

Most of the historical buildings in the area have the following characteristics:

• brick masonry of burnt clay brick and lime;

• massive perimetral walls with thicknesses of 80 cm at the basement and 40–50 cm at the top floor;
• massive longitudinal wall parallel to the main façade;
Usually, the transversal walls are thinner, 10–15 cm, without any connection with the façade walls or with a specific structural
role. They delimit the space and increase the rigidity of the building. However, there can be found structural transversal walls with
thicknesses of 30–40 cm at the staircase areas.
The horizontal structural elements above the basement are masonry vaults with thicknesses of 15–20 cm. At the same time, above
the rest of the floor, there are mostly wooden floors with a single or double layer of wooden beams [26].
All the buildings present massive slope roofs with interesting wooden frameworks and heights of 2.50–4.80 m [27]. We noticed
that they are more pronounced at the corner buildings, which is considered an urban landmark.
The height regime varies function to the buildings: starting with buildings with basement and ground floors up to buildings with
basements, ground floors, and two/ three levels [28]. The ground floor height is usually compressed between 3.20–4.40 m, while the
upper floors present a height of 3.20–3.80 m.
The total height of the buildings ranges between 4.80–15.20 m up to the starting point of the roof structure, as presented in Fig. 6.
The most common alignment is with the main façade, the longest along the street, on the longitudinal x-direction and the shortest
façade perpendicular to the street, on the transversal y-direction [2].
Historical masonry buildings are usually irregular, massive, and complex, so their structural behavior is not easy to estimate under
seismic action [29]. As far as the decay is concerned, there is some well-known typical in-plane failure mechanism for masonry
residential buildings [30,31] as presented in Fig. 7.
For the Banat seismic area, the most relevant damages can be considered those that happened during the near-field earthquake

Fig. 6. Typical structural organization of the investigated buildings according to the existing height regimes.

5
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 7. Usual in-plane failure mechanisms for masonry buildings.

back in 1991, with the epicenter in Banloc. During the local earthquake, there were registered considerable damages to historical
buildings similar to our case study. At the visual inspection, there were identified moderate decays at the lintels, arches, attics and
chimneys and not very dangerous cracks at the masonry walls and vaults. They are located in the epicenter of the seismic events, such
as Banloc, Voiteni or Ciacova, at a distance of only 30–50 km from Timisoara. These events affected especially the following
structures: vaults, arches, chimneys, towers, and roofs, as presented in Figs. 8–10 [2].
Despite the expectance of the typical failure mechanism, there can be seen that those mechanisms were activated just in a few
cases. In contrast, a specific failure mechanism can be observed from the past earthquake pictures. This particular failure mechanism,
the vertical cracks, is caused by the effect of the strong vertical forces in the epicenter of the event. This mechanism is showing that in
the case of near-field earthquakes, the vertical forces are stronger than the horizontal ones [32], so the failure mechanism that is most
likely to be activated is the vertical cracking, as presented in Fig. 11.
One of the main reasons for the appearance of this specific failure mechanism is due to the presence of surface waves in the near-
field area (Fig. 12) [32].
This type of waves, in case of the shallow earthquakes, carry the most significant amount of energy, representing the primary
cause of destruction. The waves that cause vertical cracks are the R waves that tend to move the ground up and down or sometimes
side-to-side, depending on the direction of the waves (Fig. 13a). Also, in the near-field areas, the vertical ground motions are more
important than the horizontal ones, mainly due to the appearance of the asynchrony of vertical movements (Fig. 13b). The vertical
seismic component represents, in reality, a dominant parameter in the near-field areas, even though they are usually neglected in the
seismic design [32]. Past earthquakes have revealed that many masonry typical buildings cannot deal with earthquake actions [33].
The particular in-plane failure mechanism observed in Banloc after the earthquake from 1991 is caused by the up and down
ground movement and by the asynchrony of vertical movements. Moreover, due to the high rigidity of the complex wooden fra-
mework [34], there is also activated an out-of-plane failure mechanism at the top part of the masonry wall. The entire combination of
the in-plane and out-of-plane failure mechanisms observed after the earthquake from 1991 in Banloc is described in Fig. 14.
The type of decay found after the near-field earthquake to the historical masonry buildings indicates a damage state D2-D3
(Fig. 15). This means moderate or substantial damages to nonstructural elements and small or medium damages to structural ele-
ments, without affecting the bearing capacity of the buildings [35,26].

Fig. 8. Damages occurred to massive buildings in the epicenter of past seismic events (1992).

6
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 9. Damages occurred to small buildings in the epicenter of past seismic events (1992).

Fig. 10. Diagonal shear cracking occurred to buildings in the epicenter of past seismic events (1991).

3. Case study area urban analysis

As mentioned above, Timisoara has three important historic districts. Cetate, Iosefin, and Fabric.
Even since 2012, in collaboration with the researcher prof. Mazzolani and prof. Formisano from the University of Naples, studies
were made to assess the seismic vulnerability of the buildings in Timisoara. In 2013, together with a team coordinated by Prof.
Modena, the University of Padua, the seismic vulnerability of the Cetate district was assessed [36], as illustrated in Fig. 16 [37]. As far
as the other two historical districts situated in areas outside the city fortress, Iosefin and Fabric are concerned, they are currently
studied in collaboration with the University of Naples [2,26,23]. It is worth mentioning that they were not inhabited until 1716.
Similar studies were made also in other European cities [38,39] and also in New Zeeland [40], for residential buildings or for
churches [41,42], towers [43,44] or bridges [45].
In the year 1744, there was approved the foundation of the district Iosefin, which is connected to the Cetate district by the Traian
bridge [46]. The most important buildings from the cultural, historical, and urban point of view are indicated in the map presented in
Fig. 17a. The most representative buildings have different architectural styles, i.e., the Hydro Improvement Society Palace is Neo-
classic, the Franz Marschall Palace is Art Nouveau, while the Iosefin Synagogue has an Eclectic architectural style, as shown in
Fig. 18.
The second important historical district is Fabric, connected to the Cetate district by the Decebal historic bridge, which was built
in 1910 in the Secession architectural style [46]. The most important buildings from the cultural, historical, and urban point of view
can be seen in the map presented in Fig. 17b. The most representative buildings in the Secession architectural style are the Neptun
Palace and the Countess Ana Mirbach Palace. In contrast, the Millennium Church is typical for the Eclectic architectural style, as
shown in Fig. 19.
As far as the height is concerned, in the Iosefin district, the dominant height regime is the basement + ground floor + one level,
as presented in Table 3 and Fig. 20a. Whereas for the Fabric district, most of the buildings present basement + ground floor + two
levels above, as shown in Table 3 and Fig. 20b [4]. In the figures/drawings/plans, we will use ‘B’ for the basement, ‘G.F.’ for the
ground floor, and ‘F’ the other levels.

7
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 11. Vertical cracks occurred to buildings in the epicenter of past seismic events (1991).

Fig. 12. Surface waves for near-field earthquake [32].

The urban pattern in the historical areas of Timisoara can be ranked as being of an aggregate one, i.e., that buildings form a closed
contour with a common interior yard, all the structural units work together as a group of buildings, as shown in Fig. 21 [47]. The
aggregate condition creates particular structural behavior in case of an earthquake [48].
When considering the conservation state, we noticed that, in the Iosefin district, the situation is the following: more than 75% of
the buildings are in a medium to low state, 10% are partially restored, and only 15% are recently rehabilitated (Fig. 22a).
As far as the conservation state is concerned, the situation is even worse for the Fabric district. In the case of the buildings under
our analysis, up to 82% of buildings are in a medium to low conservation state, 8% are partially consolidated and only 10% of the
buildings are recently restored (Fig. 22b) [4].

4. Empirical seismic vulnerability assessment

The specific construction typology above presented along with the actual dangerous conservation state is typical for most his-
torical buildings in Timisoara. It indicates a particular vulnerability of the historical city buildings.
To assess their seismic vulnerability, we decided to choose an Italian empirical methodology. We considered the fact that most of

8
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 13. Near-field earthquake: (a) direction of propagation for the surface waves; (b) asynchrony of vertical movements [32].

Fig. 14. The specific failure mechanism for near-field earthquakes: in-plane vertical cracking and out-of-plane overturning.

the historic masonry buildings in Timisoara date back to the Habsburgic administration, which is the same administration the
northern part of Italy was under. Therefore, there is a good similarity between the construction techniques, materials, and design
principles, as well as.
The emergency scenarios are designed to function to the existing data. Only a damage scenario obtained through a correlation
between the vulnerability, risk, the possible event intensity, and an earthquake-induced damage scale can provide valuable additional
data for the local authorities [49]. Nowadays, in many European countries, several proposals were made to design seismic vulner-
ability curves based on the registered damages after past earthquakes [50–53]. These empiric methodologies represent a statistical,
quick, and simplified way of classifying the buildings according to the most probable vulnerability level. More than this, the precise
response of the structures to the several seismic intensities is determined based on the numerical analysis. Each methodology is based
on several particularities of the investigated area and is in continuous improvement.
This paper suggests new seismic vulnerability curves for the investigated buildings in Timisoara for two seismic scenarios, which
are designed for earthquakes with the macroseismic intensity IX EMS-98. In the first stage, we used the empirical method to assess
seismic vulnerability. The number of the buildings investigated is 66 for the Iosefin district and 37 for the Fabric district.

9
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 15. Correlation between damage states and real expected damage for masonry buildings.

Fig. 16. Vulnerability map for Timisoara historical center (Cetate district).

Fig. 17. Monuments and important historic buildings: a) in Iosefin district; b) in Fabric district.

4.1. Existing Italian seismic assessment methodology

This methodology is based on a quick and simplified assessment, appropriate for the urban areas, and has two primary stages. The
first stage was first proposed by Benedetti and Petrini [54], and ten parameters are considered. It consists of the analysis of the
buildings individually, as isolated ones.
In the second stage, extra five parameters were considered to determine the influence of the adjacent buildings within the
aggregate condition. For this stage, we used the Mazzolani and Formisano methodology [55]. The final vulnerability form that was
considered is presented in Table 4.

10
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 18. Some of the most important historical buildings of Iosefin historical district.

Fig. 19. Some of the most important historical buildings of Fabric historical district.

Table 3
Height regime in Iosefin and Fabric historical districts.
Historical district Basement + ground floor Basement + ground floor + one level Basement + ground floor + 2 levels or more

Iosefin 27% 37% 36%

Fabric 10% 35% 55%

Fig. 20. The height regime analysis: a) for Iosefin district; b) for Fabric district.

The estimation of the vulnerability consists of calculating a vulnerability index Iv as the sum of the specific weighted parameters
[56]. Both stages were applied for all 103 masonry investigated buildings. The vulnerability index was determined using Eq. (5) [23]
for the buildings considered as individual isolated structural units and Eq. (6) [57] for the structures found within the entire ag-
gregate. Later on, the vulnerability index was normalized in the range of 0–100.
10
IV 10 = ∑ Si xWi
i=1 (5)

15
IV 15 = ∑ Si xWi ,
i=1 (6)

11
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 21. Type of urban pattern, aggregate structural group of buildings.

Fig. 22. Conservation state mapping for investigated buildings: (a) in Iosefin district; (b) in Fabric district.

Table 4
The vulnerability form for assessing the seismic vulnerability.
No. Factor Class Weight

A B C D

1 Organization of vertical structures 0 5 20 45 1

2 Nature of vertical structures 0 5 25 45 0.25
3 Location of the building and type of foundation 0 5 25 45 0.75
4 Distribution of plan resisting elements 0 5 25 45 1.5
5 Plain regularity 0 5 25 45 0.5
6 Vertical regularity 0 5 25 45 1
7 Type of floors 0 5 15 45 1
8 Roofing 0 15 25 45 0.75
9 Details 0 0 25 45 0.25
10 Physical conditions 0 5 25 45 1
11 Presence of adjacent buildings with different height −20 0 15 45 1
12 Position of the buildings in the aggregate −45 −25 −15 0 1.5
13 Presence and number of staggered floors 0 15 25 45 0.5
14 Effect of either structural or typological heterogeneity among adjacent structural unit −15 −10 0 45 1.2
15 Percentage difference of opening area among adjacent façade −20 0 25 45 1

where Si represents the class score, and Wi represents the weight associated with each parameter.
To be able to evaluate the possible damage states for each building and the aggregates as well, there was considered a function of
vulnerability index V, the macroseismic intensity I, and a specific Φ factor that attenuates the vulnerability curve, according to Eq. (7)
[58].
I + 6.25V − 13.1 ⎤
μD = 2.5 ⎡1 + tanh ⎛ ⎞
⎣ ⎝ Φ ⎠⎦ (7)
where ϕ was considered equal to 2.3 [52], as the buildings are predominantly residential, and V is determined following Eq. (8) [59].
IV − IVMIN
V=
IVMAX + IVMIN (8)

12
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Table 5
The correlation between damage grade, damage state, and real damage level.
µD Damage state Most probable damage level

0.0–1.5 D1 Slight (no structural damage, slight non-structural damage)

1.5–2.5 D2 Moderate (slight structural damage, moderate non-structural damage)
2.5–3.5 D3 Substantial to heavy (moderate structural damage, heavy non-structural damage)
3.5–4.5 D4 Very heavy (heavy structural damage, very heavy non-structural damage)
4.5–5.0 D5 Destruction (very heavy structural damage)

The correlation between damage grades μD , damage states (DSi), and the real level of damage to historical buildings [35] is
presented in Table 5 [26].
Based on Eq. (5), Eq. (7) and Eq. (8), there were determined the most probable damage states for each investigated building
considered as isolated structural units for macroseismic intensity IX EMS-98. The vulnerability curves for the two historical areas are
presented in Fig. 23. For the supposed macroseismic intensity, the general vulnerability of the areas is a shallow one, not precisely
following the level of decay found during the site inspection.
Based on Eq. (6), Eq. (7) and Eq. (8), there were determined the most probable damage states for each investigated building
considered in aggregate for macroseismic intensity IX EMS-98. The vulnerability curves for the aggregate condition for the two
historical areas are presented in Fig. 24. For the supposed macroseismic intensity, the general vulnerability of the areas is also a
shallow one.
The general vulnerability of the area in both previous cases was obtained as a medium vulnerability curve, showing the most
probable D1 damage state for macroseismic intensity IX EMS-98. There was used the average mean damage value for each macro-
seismic intensity. The curve is represented by a dashed line (Figs. 23–24).
The results indicate an average D1 seismic vulnerability for both historical areas, under the considered seismic scenario. The
expected damage states for all the investigated buildings is illustrated in Fig. 25 for the buildings considered as isolated structural
units, respectively, in Fig. 26 for buildings considered in aggregate.

4.2. New proposed seismic assessment methodology for near-field earthquakes

When comparing the results of the empirical Italian procedure with the real damages found after the Earthquake from Banloc in
1991, there can be seen that the empirical analysis underestimates the possible damage state. The actual damage state found on site
was D2-D3, while the expected damage state, according to the Italian empirical methodology, is D1. This difference happens because
of the local amplification conditions and the influence of the strong vertical forces.
Near-field earthquakes represent an interesting subject, due to the high impact of the strong existing vertical forces, usually
stronger than the horizontal ones. As previously shown, the near-field events cause a particular vertical cracking pattern and lead to a
damage state D2-D3 in the case of the Banat seismic region.
To adapt the vulnerability curves to the particularities of the local amplification conditions, there was proposed Eq. (9) for the
evaluation of the possible damage states.

I + 12.5V − 13.1 ⎤
μD = 2.5 ⎡1 + tanh ⎛ ⎞
⎣ ⎝ Φ ⎠⎦ (9)

where ϕ was considered equal to 2.3 [52], as the buildings are predominantly residential, and V is determined following Eqs. (5)–(8)
[59].
The new formula was obtained following real damages observed during site inspection after the earthquake from 1991 in Banloc,
Romania, and also based on nonlinear analysis results obtained by the seam research team for some of the investigated buildings [2].

Fig. 23. Vulnerability curves for first ten parameters: (a) for Iosefin historical area; (b) for Fabric historical area.

13
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 24. Vulnerability curves for all 15 parameters: (a) for Iosefin historical area; (b) for Fabric historical area.

Fig. 25. Expected damage state for the investigated buildings considered as isolated structural units: (a) Iosefin district; (b) Fabric district.

Fig. 26. Expected damage state for the investigated buildings considered in aggregate: (a) Iosefin district; (b) Fabric district.

This formula increases the damage state from D1 to D2 and even D3, as presented in Figs. 27–28 and Figs. 30–31. The most probable
damage state for each investigated building is illustrated in Fig. 29 for the buildings considered as isolated structural units, re-
spectively in Fig. 32 for buildings considered in aggregate.

14
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 27. Proposed vulnerability curves for first ten parameters for Iosefin historical area.

Fig. 28. Proposed vulnerability curves for first ten parameters for Fabric historical area.

Following Eq. (5), Eq. (6), Eq. (7) and Eq. (9), there was proposed an average empirical vulnerability curve for all the 103
buildings (Figs. 33–34). This seismic vulnerability curve is considered the general vulnerability curve for the historical districts of
Timisoara (except for the city center, where the buildings are either older, either with various consolidation work made along the
years). A vulnerability range was determined also based on the possible variability of damage (V MEC mean – 2σ; V MEC mean– σ; V MEC
mean + σ; V MEC mean + 2σ). The value of σ was determined as the standard deviation.
The study has revealed the fact that the consideration of the influence of the adjacent buildings leads to one damage state
vulnerability increasing. This effect is due to the lack of homogeneity between all the 103 investigated buildings and because the
heights of the nearby buildings are many times different.
To conclude, we can say that in case of an earthquake with IX EMS-98 macroseismic intensity, there can be expected extensive
damages to nonstructural elements. However, for the structural elements, there can appear only minor and medium damages without
affecting the bearing capacity of the buildings.
Although the structural stability of the buildings is not at risk, there should be considered the fact that all the investigated
buildings are either monuments or part of historical ensembles, sites, or protected areas. So, the damages to the nonstructural
elements could lead to losses of valuable architectural, artistic assets and irreplaceable historical values.
It is essential to highlight that the new vulnerability curves proposed are in accordance with a previous study operated through
the city center of Timisoara city together with prof. Formisano and prof. Mazzolani, which was based on the existing typological
classes (Fig. 35) [23], the classification followed the Building Typology Matrix [60].
When comparing the vulnerability curves, see Figs. 33 and 34, it can be seen a good correlation for both different methodologies.

15
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 29. Expected damage state following proposed methodology for the investigated buildings considered as isolated structural units: (a) Iosefin
district; (b) Fabric district.

Fig. 30. Proposed vulnerability curves for all 15 parameters for Iosefin historical area.

Fig. 31. Proposed vulnerability curves for all 15 parameters for Fabric historical area.

16
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 32. Expected damage state following proposed methodology for the investigated buildings considered in aggregate: (a) Iosefin district; (b)
Fabric district.

Fig. 33. Average seismic vulnerability curve and vulnerability range for all 103 buildings, for first ten parameters.

Fig. 34. Average seismic vulnerability curve and vulnerability range for all 103 buildings, for all 15 parameters.

5. Conclusion

In conclusion, when a European city is assigned to be the Capital of Culture, the understanding of its vulnerabilities is convenient
or even mandatory for the local authorities.

17
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

Fig. 35. Vulnerability curves for an aggregate located in the city center of Timisoara city.

The first part of the paper illustrates the results of an original, quick, and simplified European empirical seismic vulnerability
assessment, which indicated a shallow expected damage state.
Later on, there is presented a specific failure mechanism for historical masonry buildings in the epicenter of past earthquakes. This
particular failure mechanism illustrates a significant difference between the results of the empirical seismic vulnerability assessment
results and real damages observed after an earthquake.
The second part of the study provides a new proposed empirical seismic vulnerability assessment adapted for the near-field
earthquakes, especially for Banat seismic area. The scenario of the study considers an earthquake with the epicenter in Timisoara.
This proposed methodology is based on a new formula for the estimation of the most probable damage state. The new formula follows
the near-field earthquake effects and the real damage observed on site.
The results indicate a one or even two damage states increase for both districts, in case of an earthquake with the epicenter in
Timisoara. The increase is in comparison with the original Italian empirical seismic vulnerability assessment results.
The study proved a good correlation, for the macroseismic intensity IX EMS-98, between the new proposed methodology and
other European seismic vulnerability assessment methodologies.
The new suggested methodology might be used to assess the seismic vulnerability of any building with similar mechanical,
geometrical, architectural and urbanistic characteristics as our case study buildings, in areas with near-field earthquakes similar to
those specific to the Banat seismic area.
The next step of our study will consist of improving this proposed methodology with specific calculation and with numerical
analysis performed according to the Romanian design codes.
To conclude, we consider that for a more accurate seismic vulnerability assessment of the historical complex of buildings, the
entire aggregate has to be considered, and not only just individual isolated structural units and further numerical studies are needed.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Acknowledgments

Acknowledgments to prof. V. Stoian and D. Diaconu from Timisoara Politehnica University, Romania, for all the experimental test
reports. Also, to prof. S. Ianca and prof. M. Marin from Timisoara Politehnica University for all the information. Gratitude to
STADATA for allowing us to use the Tremuri software and to arch. A.I.Keller, arch. B. Azap and econ. C. Vasici for all their work and
collaboration.

References

[1] B. Wisner, At risk : natural hazards, people’s vulnerability, and disasters, Routledge, 2004. https://books.google.ro/books?id=drycagaaqbaj&dq=the+meaning
+of+natural+hazards&lr=&hl=ro&source=gbs_navlinks_s (accessed January 5, 2019).
[2] M. Mosoarca, I. Onescu, E. Onescu, B. Azap, N. Chieffo, M. Szitar-Sirbu, Seismic vulnerability assessment for the historical areas of the Timisoara city, Romania,
Eng. Fail. Anal. 101 (2019) 86–112, https://doi.org/10.1016/j.engfailanal.2019.03.013.
[3] E. Onescu, I. Onescu, M. Mosoarca, Multi-criterial vulnerability assessment for Timisoara city, Romania, in evaluation, in: 4th Int. Conf. Struct. Archit. ICSA,
2019.
[4] B. Azap, Management strategies for seismic vulnerability of the cultural-historical promenade of Timisoara city, Bachelor Thesis, Architecture and Urban
Planning Faculty, Politehnica University of Timisoara, 2018.
[5] N. Chieffo, F. Clementi, A. Formisano, S. Lenci, Comparative fragility methods for seismic assessment of masonry buildings located in Muccia (Italy), J. Build.
Eng. 25 (2019), https://doi.org/10.1016/j.jobe.2019.100813.
[6] S. Lagomarsino, A. Penna, A. Galasco, S. Cattari, TREMURI program: an equivalent frame model for the nonlinear seismic analysis of masonry buildings, Eng.
Struct. 56 (2013) 1787–1799, https://doi.org/10.1016/j.engstruct.2013.08.002.
[7] M. Poiani, V. Gazzani, F. Clementi, G. Milani, M. Valente, S. Lenci, Iconic crumbling of the clock tower in Amatrice after 2016 central Italy seismic sequence:
advance numerical insight, Procedia Struct. Integr. 11 (2018) (2016) 314–321, https://doi.org/10.1016/j.prostr.2018.11.041.
[8] A. Ferrante, E. Ribilotta, E. Giordano, F. Clementi, S. Lenci, Advanced seismic analyses of “apennine churches” stroked by the central italy earthquakes of 2016

18
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

by the non-smooth contact dynamics method, in: Key Eng. Mater., 2019, pp. 309–316. doi:10.4028/www.scientific.net/KEM.817.309.
[9] A. Ferrante, F. Clementi, G. Milani, Dynamic behavior of an inclined existing masonry tower in Italy, Front. Built Environ. Eng. 5 (2019), https://doi.org/10.
3389/fbuil.2019.00033.
[10] F. Clementi, A. Ferrante, E. Giordano, F. Dubois, S. Lenci, Damage assessment of ancient masonry churches stroked by the Central Italy earthquakes of, by the
non-smooth contact dynamics method, Bull. Earthq. Eng. 18 (2020) (2016) 455–486.
[11] S. Tiberti, M. Acito, G. Milani, Comprehensive F.E. numerical insight into Finale Emilia Castle behavior under, Emilia Romagna seismic sequence: damage causes
and seismic vulnerability mitigation hypothesis, Eng. Struct. 117 (2016) (2012) 397–421, https://doi.org/10.1016/j.engstruct.2016.02.048.
[12] M. Valente, G. Milani, Earthquake-induced damage assessment and partial failure mechanisms of an Italian Medieval castle, Eng. Fail. Anal. 99 (2019) 292–309,
https://doi.org/10.1016/j.engfailanal.2019.02.008.
[13] M. Valente, G. Milani, Damage assessment and collapse investigation of three historical masonry palaces under seismic actions, Eng. Fail. Anal. 98 (2019) 10–37,
https://doi.org/10.1016/j.engfailanal.2019.01.066.
[14] B. Nadolu, M. Dinca, D. Luches, Shrink Smart WP2-D4 Timisoara, Romania 2 Contents, 2010. https://www.ufz.de/export/data/400/39018_WP2_20report_
20Timisoara13639.pdf (accessed January 8, 2019).
[15] A. Szentmiklosi, A. Balarie, Contributions to the Knowledge of the Evolution of the Town of Timișoara at the End of the Middle Ages. Archaeological Preventive
Investigations within the Suburbs of Palanca Mare, Analele Ba, Timisoara, 2012. http://muzeulbanatului.ro/mbt/istorie/publicatii/ab.htm (accessed January 8,
2019).
[16] R. Radu, M. Marius, G.M. Stelian, B. Ana-Maria, Conservation and re-use of historical industrial buildings, case study fabric neighborhood, Timisoara, Romania,
Struct. Anal. Hist. Constr. 1–3 (2012), http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=ORCID&SrcApp=OrcidOrg&
DestLinkType=FullRecord&DestApp=WOS_CPL&KeyUT=WOS:000321224301101&KeyUID=WOS:000321224301101.
[17] F. Popescu, I. Ionel, C. Ungureanu, Ambient air quality measurements in Timisoara. Current situation and perspectives, J. Environ. Prot. Ecol. 10 (2009) 1–13.
[18] E. Oros, M. Popa, I.A. Moldovan, Seismological database for banat seismic region (Romania)-part 1: the parametric earthquake catalogue, 2008. www.storing.
ingv.it/es_web (accessed January 8, 2019).
[19] Ministry of regional development public administration and european funds, Romanian Design Code P100-1/2013, in Romanian, 2013. http://www.mdrap.ro/
userfiles/reglementari/Domeniul_I/I_22_P100_1_2013.pdf (accessed January 28, 2019).
[20] N. Chieffo, A. Formisano, The influence of geo-hazard effects on the physical vulnerability assessment of the built heritage: an application in a district of Naples,
Buildings 9 (2019) 26, https://doi.org/10.3390/buildings9010026.
[21] M. Kostov, Site specific estimation of cumulative absolute velocity, in: 18th Int. Conf. Struct. Mech. React. Technol. (SMiRT 18), Beijing, China, 2005.
[22] I. Apostol, M. Mosoarca, V. Stoian, Modern consolidation solutions for buildings with historical value. Part I: Reinforced Concrete Structures, Mod. Technol.
(2017) 406–413.
[23] N. Chieffo, M. Mosoarca, A. Formisano, I. Apostol, Seismic vulnerability assessment and loss estimation of an urban district of Timisoara, 3rd World Multidiscip.
Civ. Eng. Archit. Urban Plan. Symp. (2018).
[24] R. Marinov, Vulnerability of built heritage in the city of Timisoara to the seismic events, in Romanian, AGIR Report no. 4, 1999.
[25] Esteva, Geology and probability in the assessment of seismic risk, Not seen. Reported in Ambraseys (1978a), in: 2nd Int. Conf. Assoc. Eng. Geol., 1974.
[26] B. Azap, I. Apostol, M. Mosoarca, N. Chieffo, A. Formisano, Seismic vulnerability scenarios for historical areas of timisoara, in: Mod. Technol. 3rd Millenn.,
2018:, pp. 149–154.
[27] M. Mosoarca, A.I. Keller, A complex assessment methodology and procedure for historic roof structures, Int. J. Archit. Herit. 12 (2018) 578–598, https://doi.org/
10.1080/15583058.2018.1442519.
[28] I. Apostol, M. Mosoarca, N. Chieffo, A. Keller, D. Bocan, C. Bocan, R. Bradeanu, Solutions for improving seismic vulnerability of historic masonry buildings, in:
Mod. Technol. 3rd Millenn., 2018, pp. 131–136.
[29] E. Giordano, F. Clementi, A. Nespeca, S. Lenci, Damage assessment by numerical modeling of Sant’Agostino’s sanctuary in offida during the central Italy
2016–2017 seismic sequence, Front. Built Environ. 4 (2019) 87, https://doi.org/10.3389/fbuil.2018.00087.
[30] STADATA, 3muri User Manual: a computer program for analysis of structures in masonry and mixed materials through a nonlinear (pushover) and static
analysis, in: Turin, Italy, 2011.
[31] M. Valente, G. Milani, Damage assessment and partial failure mechanisms activation of historical masonry churches under seismic actions: three case studies in
Mantua, Eng. Fail. Anal. 92 (2018) 495–519, https://doi.org/10.1016/j.engfailanal.2018.06.017.
[32] V. Gioncu, F.M. Mazzolani, Earthquake Engineering for Structural Design, Spon Press, 2011.
[33] G. Milani, A. Formisano, F. Clementi, Vulnerability reduction of existing buildings and design of new structures in seismic area, AIP Conf. Proc. 1863 (2017),
https://doi.org/10.1063/1.4992610.
[34] A.I. Keller, N. Chieffo, M. Mosoarca, Influence of roof structures on seismic behavior of historic buildings, in: 3rd Int. Conf. Prot. Hist. Constr. PROHITECH’17,
2017, pp. 12–15.
[35] I. Apostol, M. Mosoarca, E. Onescu, Seismic vulnerability assessment for historical building as isolate/in aggregate for Timisoara city, Romania, in: 1st Conf.
Herit. Sustain. Innov., 2018.
[36] S. Taffarel, C. Marson, C. Valotto, M. Roverato, M. Munari, F. da Porto, C. Modena, M. Mosoarca, Seismic vulnerability maps of Timisoara historical center based
on fragility curves, SAHC, Rilem Bookseries, 2016, pp. 1605–1612.
[37] C. Valotto, S. Taffarel, C. Marson, M. Munari, F. da Porto, C. Modena, Seismic vulnerability assessment of corner buildings in the historical centre of Timisoara,
Brick Block Masonry (2016) 2499–2506.
[38] S. Lagomarsino, S. Cattari, C. Calderini, DELIVERABLE D41 European Guidelines for the seismic preservation of cultural heritage assets, 2012. https://cordis.
europa.eu/project/rcn/93579_en.html.
[39] A. Formisano, N. Chieffo, F. Fabbrocino, R. Landolfo, Seismic vulnerability and damage of Italian historical centres: a case study in the Campania region, AIP
Conf. Proc. AIP Publishing LLC, 2017, p. 450007, , https://doi.org/10.1063/1.4992616.
[40] P.B. Lourenço, D.V. Oliveira, J.C. Leite, J.M. Ingham, C. Modena, F. da Porto, Simplified indexes for the seismic assessment of masonry buildings: international
database and validation, Eng. Fail. Anal. 34 (2013) 585–605, https://doi.org/10.1016/j.engfailanal.2013.02.014.
[41] M. Valente, G. Barbieri, L. Biolzi, Seismic assessment of two masonry Baroque churches damaged by the 2012 Emilia earthquake, Eng. Fail. Anal. 79 (2017)
(2012) 773–802, https://doi.org/10.1016/j.engfailanal.2017.05.026.
[42] M. D’Amato, R. Gigliotti, R. Laguardia, Comparative seismic assessment of ancient masonry churches, Front. Built Environ. 5 (2019) 56, https://doi.org/10.
3389/fbuil.2019.00056.
[43] A. Preciado, Seismic vulnerability and failure modes simulation of ancient masonry towers by validated virtual finite element models, Eng. Fail. Anal. 57 (2015)
72–87, https://doi.org/10.1016/j.engfailanal.2015.07.030p.
[44] V. Sarhosis, G. Milani, A. Formisano, F. Fabbrocino, Evaluation of different approaches for the estimation of the seismic vulnerability of masonry towers, Bull.
Earthq. Eng. 16 (2018) 1511–1545, https://doi.org/10.1007/s10518-017-0258-8.
[45] M. D’Amato, M. Laterza, V.M. Casamassima, Seismic performance evaluation of a multi-span existing masonry arch bridge, Open Civ. Eng. J. 11 (2017)
1191–1207, https://doi.org/10.2174/1874149501711011191.
[46] Info Centrul Turistic Timisoara, in Romanian | Iosefin and Elisabetin Districts, n.d.. http://www.timisoara-info.ro/en/sightseeing/historical-quarters/iosefin/
places.html (accessed January 8, 2019).
[47] E. Onescu, I. Onescu, M. Mosoarca, The impact of timber roof framework over historical masonry structures, in: IOP Conf. Ser. Mater. Sci. Eng., 2019.
[48] M. Valente, G. Milani, E. Grande, A. Formisano, Historical masonry building aggregates: advanced numerical insight for an effective seismic assessment on two
row housing compounds, Eng. Struct. (2019).
[49] E. Quagliarini, G. Bernardini, S. Santarelli, M. Lucesoli, Evacuation paths in historic city centres: a holistic methodology for assessing their seismic risk, Int. J.
Disaster Risk Reduct. 31 (2018) 698–710, https://doi.org/10.1016/j.ijdrr.2018.07.010.

19
M. Mosoarca, et al. Engineering Failure Analysis 115 (2020) 104662

[50] R. Maio, T.M. Ferreira, R. Vicente, J. Estêvão, Seismic vulnerability assessment of historical urban centres: case study of the old city centre of Faro, Portugal, J.
Risk Res. (2016), https://doi.org/10.1080/13669877.2014.988285.
[51] G. Brando, G. De Matteis, E. Spacone, Predictive model for the seismic vulnerability assessment of small historic centres: application to the inner Abruzzi Region
in Italy, Eng. Struct. 153 (2017) 81–96, https://doi.org/10.1016/j.engstruct.2017.10.013.
[52] M. Munari, M.R. Valluzzi, G. Cardani, A. Anzani, L. Binda, C. Modena, Seismic vulnerability analyses of masonry aggregate buildings in the historical centre of
Sulmona (Italy), in: 13th Int. Conf. SFR, 2010.
[53] D. Eckhardt, A. Leiras, A.M.T. Thomé, Systematic literature review of methodologies for assessing the costs of disasters, Int. J. Disaster Risk Reduct. 33 (2019)
398–416, https://doi.org/10.1016/J.IJDRR.2018.10.010.
[54] D. Benedetti, V. Petrini, On the seismic vulnerability of masonry buildings: an evaluation method (in Italian), L’Industria Delle Costr. 149 (1984) 66–74.
[55] A. Formisano, R. Landolfo, F. Mazzolani, G. Florio, A quick methodology for seismic vulnerability assessement of historical masonry aggregates, COST Action
C26 Urban Habitat Constr. under Catastrophic Events, 2010, https://doi.org/10.13140/2.1.1706.3686.
[56] V.I. Novelli, Hybrid Method for the Seismic Vulnerability Assessment of Historic Masonry City Centres, Thesis, 2017.
[57] A. Formisano, N. Chieffo, M. Mosoarca, Seismic vulnerability and damage of a historical centre in the seismic vulnerability and damage of a historical centre in
the district of Caserta (Italy), in: 3rd Int. Conf. Prot. Hist. Constr. PROHITECH’17, 2017.
[58] R. Vicente, S. Parodi, S. Lagomarsino, H. Varum, J.A.R. Mendes, D. Silva, Seismic vulnerability assessment, damage scenarios and loss estimation. Case study of
the old city centre of Coimbra, Portugal, in: 14th World Conf. Earthq. Eng., 2008.
[59] R. Maio, Seismic Vulnerability Assessment of Old Building Aggregates, Universidade de Aveiro, 2013.
[60] P. Moroux, B. Le Brun, Presentation of RISK-UE project, Bull. Earthq. Eng. 4 (2006) 323–339.

20