Comprehensive Material Characterization and Seismic Performance Evaluation of a Traditional Masonry Residential Building with Reinforced Concrete Slabs

Abstract

Reinforced concrete began replacing traditional masonry construction in the early 20th century, yet hybrid buildings combining unreinforced masonry (URM) walls with concrete slabs remain prevalent in Istanbul. Understanding their seismic behavior is critical for risk mitigation and heritage preservation. This study investigates a seven-story masonry residential building with cast-in-place reinforced concrete slabs constructed in 1953. The assessment involved non-destructive inspections, double flat-jack and shear tests, and geophysical site surveys. A finite element model was developed using Midas Gen software v2020 and analyzed through linear response spectrum and nonlinear pushover analyses based on TBSC-18 and SRMGHS-17. The modulus of elasticity ranged from 200.2 MPa to 1062.2 MPa, and bed joint shear strength varied between 0.50 MPa and 0.79 MPa. The building satisfied inter-story drift criteria for limited damage (SL-3), controlled damage (SL-2), and pre-collapse (SL-1). However, it failed to meet the shear force requirements at all levels. Pushover analysis revealed ultimate lateral capacities of 11,997 kN in the x-direction and 16,209 kN in the y-direction. The findings highlight the shear vulnerability of such hybrid systems and underscore the importance of combining experimental characterization with numerical modeling to develop effective retrofitting strategies.

1. Introduction

In Turkey, masonry structures were once a widely used construction technique. However, due to the presence of three major active fault lines and the occurrence of destructive earthquakes, this construction method has gradually been abandoned, especially in major cities like Istanbul. Nevertheless, the adoption of reinforced concrete—a new construction method with significantly higher ductility under seismic effects compared to unreinforced masonry (URM)—was not a rapid transition in the early 20th century. During this transition period, structures were designed to combine masonry walls with reinforced concrete elements.

Chieffo et al. assessed Muccia’s seismic vulnerability after the 2016 earthquake, comparing empirical and mechanical fragility methods. They evaluated masonry building aggregates using the Building Typology Matrix (BTM) and an index-based approach. A damaged masonry building case study compared empirical fragility curves with nonlinear static analysis using NTC18-defined mechanical properties at a limited knowledge level (LC1) [1]. Casapulla et al. analyzed corner failure in masonry buildings using pushover and dynamic analyses. The study incorporated a displacement-based static method and an updated dynamic model to assess seismic response. A case study on the 2016–2017 Central Italy earthquakes validated the models, showing substantial predictive accuracy. The findings highlighted the static model’s conservatism and effectiveness in evaluating rocking masonry structures [2]. D’Altri et al. reviewed and categorized computational modeling strategies for masonry structures, emphasizing their seismic vulnerabilities despite their effectiveness in handling gravity loads. The study classified these models into the following four main types: block-based, continuum, geometry-based, and macro element approaches, offering a structured overview of their applications and limitations. While recognizing the challenges in standardizing these methods, the research clarified existing methodologies and discussed potential advancements in computational analysis for seismic assessment [3]. Ravichandran et al. evaluated approaches for the earthquake analysis of non-engineered masonry constructions (NECs), emphasizing computational efficiency and accuracy. They analyzed a representative NEC using 3D ANSYS 2015 models and a simplified 2D SAP2000 model using static and dynamic loads. The findings indicated that the simplified model provided a reliable and efficient alternative for assessing and mitigating seismic risk in NECs [4]. Bidaj et al. investigated the earthquake performance of URM structures by analyzing a template building severely damaged during the 2019 Albanian earthquakes. They employed a macro-modeling approach to evolve a three-dimensional finite element model (FEM) using material properties obtained from experimental tests. Pushover analyses assessed the building’s structural response and identified failure mechanisms. The findings suggested that existing predictive models often overestimated lateral resistance, while the observed damage closely matched numerical predictions. This study provided valuable insights for improving seismic risk assessments and retrofitting strategies for URM buildings in earthquake-prone areas [5]. Valluzzi et al. examined the effects of tan earthquake in Italy that caused destruction to 2300 buildings and affected 20 villages. The study investigated the relationship between local macro seismic intensity and the extent of damage. To evaluate their impact on structural performance, the researchers compared various factors. Based on their findings, they developed new methods for assessing seismic vulnerability, including statistical evaluations, typological classifications, and construction parameters [6]. Sarhosis et al. investigated the structural damage to homes, schools, and churches in the Thessaly region of Greece caused by earthquakes. The study found that, whereas historically built brick buildings suffered severe damage, those designed according to the most recent seismic norms suffered minor damage. The findings highlighted the necessity of thoroughly evaluating these structures in rural areas to reduce future seismic risks [7]. Morandi et al. evaluated the in-plane seismic performance of the “Resisto 5.9” steel modular reinforcement system designed for unreinforced masonry (URM) walls. This study incorporated experimental and numerical analyses, including full-scale cyclic tests and Distinct Element Method (DEM) simulations, to assess key performance metrics such as lateral strength, stiffness, and displacement capacity. The results indicated that the reinforcement system significantly enhanced displacement capacity while preserving lateral strength, thereby improving seismic strengthening techniques for URM structures in earthquake-prone regions [8]. Blash et al. investigated the compressive performance of URM walls, focusing on design provisions, experimental research, and key influencing factors such as slenderness ratio, axial load, and openings. The study identified inconsistencies in existing masonry design standards and emphasized the need for full-scale testing to enhance predictive models. Through a critical analysis of experimental findings and construction methods, the research provided insights to refine masonry design provisions and improve the understanding of URM wall behavior under compression. This work offered guidance for future research and structural assessments [9]. Buonocunto et al. developed an analytical fragility model for unreinforced masonry (URM) buildings in Italy’s Campania region, which was significantly impacted by the 1980 Irpinia earthquake. They employed the equivalent frame approach with nonlinear macro-elements to assess seismic fragility through static analysis. This model, validated against the damage caused by the Irpinia earthquake, enhanced regional seismic risk assessments, and strengthened national vulnerability frameworks [10]. Doran et al. evaluated numerical modeling techniques for unreinforced masonry (URM) walls through experimental tests on specimens with and without openings subjected to lateral and axial loads. They implemented macro- and detailed micro-modeling approaches in ABAQUS and compared the results to the experimental data. The findings demonstrated that both methods effectively simulated the in-plane response of URM walls, providing reliable tools for structural assessment [11]. Dedeoglu et al. analyzed the structural effects of the Düzce–Gölyaka earthquake, examining masonry structures in suburban settlements. The findings indicated that inadequate engineering oversight and poor construction quality were the primary factors contributing to structural damage [12]. Tripathy and Singhal examined how openings influence the in-plane load capacity of URM walls through experimental and numerical analyses. Testing four half-scale URM walls under cyclic loading revealed significant damage around the openings, with door openings reducing the capacity by over 25%. Nonlinear finite element analysis indicated capacity reductions between 6% and 82%, depending on the size and location of the openings and material properties. They proposed a validated simplified equation and simulated the in-plane behavior of masonry, providing insights into the seismic vulnerability of perforated URM walls and improving structural assessment methods [13]. Caicedo et al. evaluated the seismic vulnerability of historical masonry structures through two case studies simulated in OpenSees, incorporating nonlinear wall interactions and accounting for European seismic hazards. The fragility functions derived from incremental dynamic analysis (IDA) were compared with those obtained through probabilistic seismic demand analysis (PSDA). The results demonstrated that IDA provided more conservative predictions of collapse. These findings underscored the significant impact of record-to-record variability and modeling uncertainties on seismic vulnerability assessments [14]. Keshmiry et al. provided a thorough review of assessing, repairing, and retrofitting masonry structures, emphasizing their vulnerabilities and preservation challenges. The study examined various damage detection techniques and evaluated retrofitting methods for their effectiveness in improving structural resilience. Case studies demonstrated practical applications, offering insights to improve masonry assessment techniques and retrofitting strategies [15]. Dauda et al. analyzed 54 studies to identify inconsistencies in testing methodologies, specimen characterization, and loading conditions, highlighting the need for standardized protocols. Additionally, the study identified significant research gaps, particularly regarding the effects of openings and the interactions between masonry walls and structural elements. The authors recommended developing uniform testing procedures and encouraging further experimental and numerical investigations to enhance our understanding of URM wall behavior [16]. Ahn et al. developed seismic fragility curves for Korea’s unreinforced masonry (URM) buildings by using incremental dynamic analysis to assess earthquake damage and recovery needs. They modeled eight different URM building types, considering wall openings and performance-based damage criteria. The study analyzed how the intensity of ground motion affects the probability of collapse. The resulting fragility curves were validated using damage data from the 2017 M5.4 Pohang earthquake, ensuring alignment with Korean post-earthquake reconnaissance guidelines [17]. Najafi et al. investigated the reliability of vulnerability curves derived from incremental dynamic analyses by applying the macro-modeling approach in OpenSees to an unreinforced masonry (URM) school building. The findings indicated that drift-sensitive demands showed lower variability, while acceleration-sensitive demands exhibited more significant variability at higher intensity levels. Additionally, the likelihood of collapse closely aligned with the thresholds established by seismic design codes. These results highlighted the effectiveness of the proposed methodology in assessing seismic risk and guiding retrofitting strategies for URM structures in seismically active areas [18]. Gautam et al. assessed the structural condition of neoclassical monuments in Nepal using ambient vibration measurements and system identification techniques. Given their seismic vulnerability and historical importance, these structures require regular evaluation. By analyzing vibration frequencies before and after retrofitting, the study tracked changes in stiffness, providing valuable insights into damage progression and the improvements made to the structures. The findings highlighted the effectiveness of non-invasive monitoring in assessing seismic vulnerability and evaluating the retrofitting of heritage masonry buildings [19]. Bilgin et al. conducted field surveys and carried out equivalent frame pushover analyses on seven representative 1970s Albanian masonry buildings that had been damaged during the 2019 Mw 6.4 Durrës earthquake sequence. They determined that the unreinforced, poorly anchored wall systems had reached a Near-Collapse state at a peak ground acceleration (PGA) of roughly 0.16 g and had collapsed at PGA levels of 0.26 g or higher—thresholds that lay below the region’s 475-year design hazard of 0.30–0.40 g. Their results indicated that slender walls, weak mortar, and inadequate diaphragm ties had constituted the principal vulnerabilities, and they concluded that the existing masonry stock had been unable to meet the foreseeable seismic demands and, therefore, required urgent retrofit interventions [20]. Pouraminian developed a nonlinear finite element macro model of the 16.11 m Gaskar brick minaret and combined it with a probabilistic load resistance assessment, in which masonry stiffness and strength had been Monte-Carlo-sampled at a 10% coefficient of variation. The analysis showed that the minaret satisfied both code-level and extreme wind demands. However, its computed seismic reliability index, derived from peak displacement limits under three strong ground motions, indicated vulnerability to slight earthquake-induced damage [21]. Kocaman and Gürbüz developed finite element models of three 15th- to 16th-century single-domed mosques in Erzurum and subjected them to ground motions recorded during the 1992 Erzincan, 1997 Düzce, and 2023 Kahramanmaraş earthquakes. The analyses identified weak spots at wall junctions and dome interfaces; moreover, after the researchers introduced discreet diaphragm ties and dome-ring stiffening, the models showed sharply reduced displacements and damage. The results indicated that these low-impact retrofits could reliably protect similar historic mosques against future earthquakes [22]. Onat and Usta Evci conducted a finite element parametric study of a reinforced concrete building that had been damaged in the 2020 İzmir earthquake, modeling it in SAP2000 V24 first as a bare frame and then as frames infilled with Eurotherm clay blocks, burnt clay brick, and autoclaved aerated concrete (AAC) blocks. Their nonlinear time–history analyses with seven Turkish earthquake records showed that infilled frames exhibited shorter fundamental periods, higher ground-floor and global base shear forces, greater column axial load capacities, and lower inter-story drift ratios than the bare frame. Porotherm and burnt-brick infills delivered similar stiffness and strength gains. In contrast, AAC infills imparted markedly lower stiffness, highlighting the necessity of explicitly considering infill wall material properties in seismic assessment and retrofitting reinforced concrete buildings [23].

In addition to studies evaluating structural performance, the literature focused on material characterization through testing. Alecci et al. examined the influence of masonry wall thickness on the accuracy and outcomes of double flat-jack tests. They performed full-scale tests on walls of varying thicknesses to evaluate their deformability properties and elastic modulus. Following these tests, simulations were conducted to analyze thicker wall configurations. The findings revealed that the surrounding masonry and wall thickness significantly affected the test outcomes, with thicker walls potentially overestimating the elastic modulus by as much as 40%. An empirical correction factor was introduced to address this issue, which provided more accurate assessments of masonry deformability and improved the reliability of existing testing standards [24]. Segura et al. compared triplet and couplet test configurations, addressing the challenges associated with triplet tests, including asymmetric failure and interpretation difficulties. They conducted a comparative experimental program on two types of low-strength masonry, including a composition resembling historical materials, to assess Mohr–Coulomb parameters, residual shear strength, and fracture energy. The findings revealed that couplet specimens consistently provided higher estimates of these parameters, demonstrating their reliability as an alternative testing method. This research improved the methodologies for assessing masonry shear strength and contributed valuable data on the properties of historic masonry [25]. Pirchio et al. conducted rebound hammer and pulse velocity tests on 170 specimens collected from 72 churches, categorizing the masonry into four types of unreinforced masonry (URM). Using the SonReb approach, they developed predictive equations to estimate the mechanical properties, calibrated them with expert judgment, and partially validated them against modern URM specimens. The study highlighted non-destructive testing (NDT) as a quick and objective assessment tool for URM properties and called for further research to improve predictive accuracy [26]. Avila-Haro et al. conducted a structural condition assessment of critical neoclassical monuments in Nepal using system identification techniques to evaluate their seismic vulnerability and the effectiveness of retrofitting. Due to their historical and administrative significance, these buildings required periodic evaluations to ensure safety and resilience. The study involved ambient vibration measurements taken from six monuments in the Kathmandu Valley, analyzing changes in vibration frequencies before and after retrofitting to assess variations in stiffness. The findings indicated that non-invasive structural monitoring effectively captured the progression of damage and the impact of retrofitting [27]. Lulić et al. outlined the flat-jack method’s principles, challenges, and potential advancements, providing practical recommendations to improve its reliability. The findings contributed to assessing masonry structures in Mediterranean Europe and supported numerical modeling and retrofitting strategies to improve seismic resilience [28]. Stepinac et al. assessed the impact of the 2020 Croatia earthquakes on URM constructions, attributing significant damage to material degradation, insufficient preservation, and outdated seismic design. The study focused on the flat-jack method to evaluate masonry’s mechanical characterization through stress transfer analysis. The findings improved the understanding of Mediterranean brick masonry’s mechanical behavior, enhancing structural assessment accuracy and informing reconstruction strategies [29]. Łątka investigated the influence of flat-jack geometry and spacing on strain measurement accuracy during laboratory testing of brick masonry walls. The study analyzed three flat-jack shapes and spacings of three, five, and seven brick layers. The findings revealed that semi-circle flat-jacks overestimated wall stiffness because of their limited surface area and shallow placement within the wall bed joint. The research demonstrated that double flat-jack tests offered a minimally invasive method for evaluating deformation parameters, with recorded values being comparable to those reported in previous research [30]. Parsekian et al. centered on validating a novel flat-jack method to assess hollow concrete block masonry. This flat-jack acted as a non-destructive testing tool, deriving stress–strain relationships and estimating masonry strength on-site. While traditional flat-jack techniques were primarily suitable for solid masonry units, such as bricks, this study customized the method to accommodate hollow block applications. The experimental assessments of existing masonry structures demonstrated the device’s effectiveness in estimating the compressive strength and elasticity modulus while minimizing structural disruption. Comparative results from laboratory tests confirmed the reliability of this method, highlighting the practicality and precision of the flat-jack for evaluating modern masonry systems [31]. Thamboo et al. explored the challenges of assessing non-uniform stone construction in historic buildings and developed an experimental database including compression and shear test results. The study classified masonry, analyzing mechanical properties before and after reinforcement. Additionally, it evaluated traditional and innovative retrofitting methods, providing insights into masonry strengthening [32]. García et al. presented a macro-modeling approach to assess the effective material characteristics of masonry structures, supported by experimental verification. The study tackled the challenges of modeling masonry walls within a 3D framework by deriving effective properties through numerical simulations on representative volume elements (RVEs). This approach demonstrated its effectiveness in improving the accuracy of macro-modeling techniques for evaluating the seismic performance of masonry structures [33].

The evaluation of seismic performance in traditional masonry buildings remains a critical area of research, particularly in earthquake-prone regions. Given masonry materials’ inherent variability and low tensile strength, such structures often demonstrate inadequate seismic resilience, highlighting the necessity for comprehensive assessment methods. Researchers have increasingly used finite element analysis (FEA) in recent decades to analyze and forecast how masonry structures would respond to seismic loads. This study focuses on a traditional masonry building featuring reinforced concrete slabs and employs a multi-stage investigation approach to characterize the materials and assess their seismic performance. The methodology encompasses in situ material inspections, employing non-destructive and semi-destructive methods, including the double flat-jack test, shear tests, and geotechnical investigations, to analyze subsoil conditions. Using the data gathered, a finite element model was developed, and linear and nonlinear assessments were carried out using the Turkish Building Seismic Code [34] and the Seismic Risk Management Guide for Historical Structures [35]. Three earthquake scenarios—SL-3, SL-2, and SL-1—with associated exceedance probabilities of 50%, 10%, and 2%, respectively, were used to assess the seismic performance. This multi-phase study guides the evaluation of traditional masonry buildings’ performance.

2. The Examined Building

The Kadırgalar Apartment is a multi-story residential building with a basement, ground floor, five further stories, and an attic in Istanbul’s Şişli district. The architectural design was completed in 1953 by İbrahim Rana Zıpçı, a prominent architect in Turkey, with construction starting during the same year. The building is nearly rectangular, with a width of 12 m at the front and back façades and side façades extending 25 m in length. Its primary structural system consists of load-bearing masonry walls, while the floor slabs are reinforced concrete. The basement features a stone masonry wall system designed to withstand soil pressures, but the main load-bearing structure is constructed with solid clay bricks. The masonry walls are 50 cm thick in the basement, 40 cm thick on the ground floor, and about 30 cm thick on the higher stories. Given the building’s construction era and its current condition, the concrete used in its structure was determined to be cast-in-place. Based on the examination of the observation pit excavated on-site, it has been determined that the structure’s foundation type is a continuous footing. The porous nature of the concrete, inconsistencies in aggregate gradation, and the presence of visible marine shell fragments reflect the materials and construction techniques typical of that time. Currently functioning as a residential building, the Kadırgalar Apartment is a significant example of mid-20th-century masonry construction in Istanbul, showcasing the architectural style and material characteristics. Figure 1 presents older images of the structure, while Figure 2 illustrates its location and images of all four façades. Figure 3 provides floor plans for the basement, ground floor, and upper floors, with the first floor plan as a representative example. In addition to the general architectural layout and material configuration, the structural investigation revealed further details of the load-bearing system. The floor system of the building consists of cast-in-place reinforced concrete slabs, providing rigid diaphragm behavior at each story level. There are no reinforced concrete cores at the corners of the building or embedded within the masonry walls. Furthermore, the door and window openings are not framed with reinforced concrete elements. These observations were confirmed through on-site inspections and localized removal of interior plaster. These characteristics were considered in the structure’s finite element modeling.

3. Material Tests

Figure 4 presents a flowchart illustrating the material tests, the structure’s finite element modeling, and the analysis techniques used. On-site research has been carried out to identify the materials utilized in the building. As seen in Figure 5, plastered walls were partially stripped to observe the structural materials. The analysis indicated that the building material primarily consists of brick, with roughly hewn stone masonry found only in the basement. As depicted in Figure 6, comprehensive material characterization was undertaken through a combination of in situ visual surveys and semi-destructive tests. The masonry walls are built of solid fired-clay bricks with nominal dimensions of 210 mm × 100 mm × 70 mm (length × depth × height). Both the bed (horizontal) and head (vertical) mortar joints exhibited an average thickness of approximately 15 mm. Double flat-jack and in situ shear tests produced elastic modulus values ranging from 200.2 MPa to 1062.2 MPa and bed joint shear strengths between 0.50 MPa and 0.79 MPa. Figure 7 illustrates the spatial distribution of the flat-jack and shear test locations, and the experimentally derived parameters were directly adopted for assigning material properties in the numerical model.

3.1. Flat-Jack Tests

The walls underwent double flat-jack tests. Six distinct building walls were subjected to double flat-jack tests conducted using ASTM C1197 [36], using a digital flat-jack-testing system (model 30-WF6016) (Figure 8). Before the tests, a testing area containing at least five rows of bricks was identified on each wall element. The mortar joints above and below this area were removed, and flat-jack plates were inserted into the prepared slots. To measure the vertical deformations in the region between the two plates, Linearly Variable Differential Transformers (LVDTs) with electronic instrumentation were installed vertically on the wall surfaces (Figure 9). The compressive load was applied to the plates using a hydraulic pump with an oil-pumping mechanism. A data collection device recorded the stress and strain values generated in the walls. Load and displacement transducers were connected to the datalogger, and all sensor signals were captured with the manufacturer’s proprietary data acquisition DATACOMM 2 software. Based on the collected data, stress–strain curves were created for each wall. Using these data, the elastic modulus (E, MPa) of the composite walls was calculated, and the results are presented in

Table 1. Findings for the shear strength index and modulus of elasticity for structural components.

Floor	Modulus of Elasticity (E, MPa)	Shear Strength Index (τ, MPa)
ground	1058	0.79
first	464.6	0.50
second	646.9	0.74
third	435.6	0.66
fourth	1062.2	0.78
fifth	200.2	0.53

. Equation (1) states that the stress in the brickwork between the flat-jacks at any given location (f_m) is determined by the flat-jack pressure (p) in MPa. This value is multiplied by a dimensionless constant (K_m) representing the flat-jack’s stiffness and geometrical characteristics. The calculation of the elastic modulus (E) involves the ratio of an increment of stress (δ_fm) to the corresponding increment of strain (δε_m).

$${\mathrm{f}}_{\mathrm{m}}={\mathrm{K}}_{\mathrm{m}}\mathrm{ p}$$

(1)

$$\mathrm{E}={\displaystyle \frac{\mathsf{\delta }{\mathrm{f}}_{\mathrm{m}}}{\mathsf{\delta }{\mathsf{\epsilon }}_{\mathrm{m}}}}$$

(2)

3.2. Shear Tests

Shear tests were conducted on the load-bearing masonry walls located on the ground and typical stories in accordance with ASTM C1531 Method B [37], using a digital shear-testing system. Figure 10 illustrates the shear test setup according to this standard. Six different locations were selected—one on each floor (Figure 11). Method B does not employ flat-jacks to regulate the normal compressive stress, thereby maintaining the natural stress state of the wall during testing. The horizontal load applied to each panel was measured with a calibrated load cell, while the corresponding displacement was recorded with an LVDT. Data from both sensors were logged synchronously. Figure 12 depicts the internal stress distribution in the panel during the application of the horizontal load. The maximum horizontal force resisting slip (P_h) was divided by the net bed joint contact area (A_j) to calculate the bed joint shear strength index (τ). Because all panels were tested beneath window openings where vertical loading is negligible, the normal stress (σ_b) was taken as 0 MPa, so the initial shear strength (τ₀) equals τ. Each test was terminated when continuous slip became evident.

$$\mathsf{\tau }={\displaystyle \frac{{\mathrm{P}}_{\mathrm{h}}}{{\mathrm{A}}_{\mathrm{j}}}}$$

(3)

$${\mathsf{\tau }}_0=\mathsf{\tau }-\mathsf{\mu } \mathrm{\ast } {\mathsf{\sigma }}_{\mathrm{b}}$$

(4)

A statistical evaluation of the material test results was conducted to better quantify the variability in mechanical properties. The masonry walls’ average elastic modulus was 620.1 MPa, with a coefficient of variation (CoV) of 23.5%. The bed joint shear strength had an average value of 0.64 MPa, with a CoV of 18.2%. These results indicate a moderate level of material heterogeneity, typical for aged masonry structures.

4. Soil Tests

Geophysical investigations were conducted to characterize the subsurface soil conditions and evaluate the soil–structure interaction (SSI) effects influencing the seismic performance of the building. Seismic refraction and Multichannel Analysis of Surface Waves (MASW) techniques were employed. These methods installed geophones by embedding them into soft soils or using specialized apparatus to maintain vertical orientation on hard surfaces such as rock, asphalt, concrete, or mosaic-tiled surfaces. This study conducted seismic tests on a mosaic-tiled floor, with geophones coupled adequately to the surface and the impact source adjusted accordingly to ensure reliable data acquisition. Figure 13 shows the setup of the seismic test. While a rigid surface may affect very shallow wave propagation characteristics, it does not significantly influence the deeper shear wave velocity profiles critical for engineering evaluations. P-wave and S-wave velocities were measured to determine the mechanical and dynamic properties of the subsurface layers. Surface waves, which propagate along the ground surface and are highly sensitive to near-surface material properties, provided critical input for constructing the site’s seismic velocity profile [38].

Table 2. Parameters obtained from seismic tests.

Parameters	Unit	Layer 1	Layer 2
Vp	m/s	2200	3300
Vs	m/s	760	987
Average thickness	m	2	-
Vp/Vs	-	2.89	3.34
Poisson’s ratio [39]	-	0.43	0.45
Density [40]	gr/cm3	2.19	2.31
Shear modulus [41]	kg/cm2	12,675	22,459
Elasticity modulus [39]	kg/cm2	36,308	65,172
Bulk modulus [39]	kg/cm2	89,313	221,122
Bed coefficient	t/m3	7148	9752
Bearing capacity [42]	kg/cm2	8.34	11.38
Design strength	kg/cm2	5.96	8.13
Fundamental period of soil	s	0.21	0.21
Vs30	m/s	1236	1236
Soil amplification factor	-	1	1

lists the parameters used in the geotechnical investigations, whereas

Table 3. Simplified soil stratigraphy at the test location.

Layer No.	Depth Range (m)	Soil Description
1	0–2.0	medium-dense silty clay
2	>2.0	dense sand/weathered rock

presents the soil classifications obtained from the tests as a function of depth. The MASW data indicated significant variability in shear wave velocities, ranging from 190 m/s to 420 m/s across different strata. These measurements were utilized to define soil stratification, estimate the thickness of individual layers, and inform boundary condition modeling within the finite element analysis. This approach realistically incorporated the interaction between the basement structure and the surrounding soil, ensuring a more accurate seismic performance evaluation.

5. Finite Element Method

In this study, the reinforced concrete slabs of the investigated building were explicitly modeled with shell elements in MIDAS Gen. Each slab was assigned its actual thickness and material properties (elastic modulus, Poisson’s ratio, and density) together with realistic boundary conditions. By adopting shell elements, both the mass and the in-plane/out-of-plane stiffness contributions of the slabs were fully captured. To reproduce the rigid diaphragm behavior expected under seismic loading, diaphragm constraints were imposed at every floor level so that all nodes within a story were translated together in the horizontal plane. This modeling strategy ensures an accurate representation of the rigid horizontal diaphragm effect and the interaction between slabs and masonry walls in the seismic performance evaluation.

The seismic performance of the masonry apartment building under consideration was evaluated using two seismic guidelines: TBSC-18 [34] and SRMGHS-17 [35]. The finite element model of the structure represents masonry walls as shell elements with a 50 cm mesh size. In the finite element modeling of masonry structures, the selection of mesh size is critical for balancing computational efficiency with the accuracy of results. Previous studies [43,44] suggest that, for thick-walled masonry structures, a mesh size approximately equal to the wall thickness is sufficient to ensure reliable global behavior while maintaining reasonable computational demands. Accordingly, in this study, a mesh size of 50 cm was selected, considering the typical wall thicknesses of the structure (ranging from 30 cm to 50 cm). This choice accurately captured global deformation patterns without excessively increasing computational cost. Additionally, Lourenço [43] emphasized that maintaining a mesh size comparable to the structural element thickness in shell element modeling of masonry structures provides adequate precision for simulating the seismic behavior of traditional thick-walled buildings. The modeling approach reflects the building’s structural characteristics, particularly the presence of monolithic reinforced concrete floor slabs. These slabs were assumed to act as rigid diaphragms, ensuring complete moment transfer to the masonry walls. The structure, which consists of 29,030 nodes and 31,369 elements, was modeled using the finite element software Midas Gen v2020 [45]. Figure 14 displays the front and rear elevation views of the building’s finite element model. A macro-modeling strategy was adopted, whereby each masonry wall was idealized as an equivalent homogeneous orthotropic shell. This approach captures the global stiffness, mass, and strength of the masonry without explicitly discretizing individual bricks and mortar joints, following the recommendations of Lourenço [43] and D’Altri et al. [3]. The live load of 2.0 kN/m² and the death load of 1.5 kN/m² were considered for the slabs. Based on the model, the software automatically calculated the building’s self-weight and incorporated it into the study. A damping ratio of 5% was applied, consistent with the typical values for masonry structures recommended in seismic design guidelines [34,46,47]. Furthermore, the adoption of a 5% damping ratio in the dynamic analysis of historical masonry buildings is well supported by recent case studies. In the seismic assessment of the “Pietro Capuzi” School in Visso, Marche Region, Italy, a 5% damping ratio was applied to reproduce the building’s response to earthquake loading [44]. Similarly, AlShawa et al. [48] employed a 5% Rayleigh damping coefficient for all materials in their finite–discrete element simulations of a half-scale stone masonry aggregate, achieving good agreement with shake-table test. These applications illustrate the widespread practice of assigning a 5% damping ratio to historical masonry structures in seismic analyses. This value accounts for energy dissipation through non-structural components, such as mortar joints, which influence the overall dynamic behavior of the structure.

5.1. Linear Analysis

The Earthquake Hazard Map [49] was used to create elastic spectrum graphs that included the building’s location and the soil classification (ZB; slightly weathered, moderately strong rocks). The calculations for design spectral acceleration coefficients utilized in developing these graphs are detailed in

Table 4. Calculation of seismic parameters.

Explanation	Formulation
Short-period design spectral acceleration coefficient	${\mathrm{S}}_{\mathrm{D}\mathrm{S}}={\mathrm{S}}_{\mathrm{S}} {\mathrm{F}}_{\mathrm{S}}$
Design spectral acceleration coefficient for a 1.0-s period	${\mathrm{S}}_{\mathrm{D}1}={\mathrm{S}}_1 {\mathrm{F}}_1$

S_s, S₁: spectral acceleration coefficients; F_s, F₁: local site effect coefficients

.

Table 5. Soil and earthquake parameters.

Parameters	Unit
soil class	ZB *
seismic ground motion level	SL-3, SL-2, SL-1
earthquake map spectral acceleration coefficients	SL-3, Ss = 0.333, S1 = 0.096 SL-2, Ss = 0.841, S1 = 0.236 SL-1, Ss = 1.475, S1 = 0.412
peak ground acceleration	SL-3, PGA = 0.145 g SL-2, PGA = 0.346 g SL-1, PGA = 0.595 g
local soil coefficients	SL-3, Fs = 0.900, F1 = 0.800 SL-2, Fs = 0.757, F1 = 0.800 SL-1, Fs = 0.900, F1 = 0.800
spectral acceleration coefficients	SL-3, SDS = 0.300, SD1 = 0.077 SL-2, SDS = 0.757, SD1 = 0.189 SL-1, SDS = 1.327, SD1 = 0.330
live load participation coefficient (n)	0.30

* Moderately weathered, medium-strong rocks.

presents the soil and earthquake parameters used in the generation of the elastic spectrum. Using SRMGHS-17 [35], the following three seismic levels were considered: SL-3, SL-2, and SL-1. Figure 15 illustrates elastic spectrums for the earthquake load reduction coefficient (Ra) set to 1. The applied base shear forces along the x- and y-axes were greater than 80% of the equivalent earthquake load, according to the Turkish Building Seismic Code [34].

Table 6. The numerical model’s material properties [34].

Material	Elasticity Modulus (MPa)	Material Type	Weight Density (N/mm3)	Poisson’s Ratio
brick walls	750	Isotropic	2.2 × 10−5	0.25
stone walls	3000	Isotropic	1.8 × 10−5	0.25
reinforced concrete	24,000	Isotropic	1.2 × 10−5	0.25

outlines the material properties incorporated into the finite element model. The values listed in this table were derived using the reference tables provided in the TBSC-18 [34] guideline.

For displacement control evaluations, the examined structure’s relative story drifts are calculated. SRMGHS-17 [35] states that a structure satisfies the “limited damage” performance level if the relative story drift for SL-3 is less than 0.3%. It satisfies the “controlled damage” performance level if it does not surpass 0.7% for SL-2. It meets the “pre-collapse” performance criterion if it is less than 1.0% for SL-1.

Table 7. Parameters used in response spectrum analyses [31].

Parameters	Unit
evaluated performance goals and ground motion levels during earthquakes	LD for SL-3 CD for SL-2 PC for SL-1
limit values for story drift	0.3% for LD 0.7% for CD 1.0% for PC
local soil coefficients	SL-3, Fs = 0.900, F1 = 0.800 SL-2, Fs = 0.757, F1 = 0.800 SL-1, Fs = 0.900, F1 = 0.800
coefficient of earthquake load reduction, Ra	Ra = 1 for SL-3 Ra = 3 for SL-2 Ra = 3 for SL-1

LD: limited damage, CD: controlled damage, PC: pre-collapse.

summarizes this information. The window margins—among the most vulnerable parts of the masonry walls—were designated as the crucial sections to assess whether the building satisfies the shear force performance requirements (Figure 16). In the shear force performance assessment, in accordance with TBSC-18 [34], the ratio of the adverse shear force in a given story to the base shear force must not exceed 40%. The applied earthquake load combinations include dead and live loads, and for each direction, both positive and negative loading cases were considered, resulting in a total of four loading directions along the x- and y-axes. The Ritz vectors approach was used to perform modal analysis to assess the dynamic properties of the structure under investigation. Figure 17 presents the first four mode shapes of the building, while

Table 8. Results of modal analysis.

Mode	Period (s)	Mass-Participation Direction	Mass-Participation Ratios (%)
1	0.795	y	49.4
2	0.719	y	23.9
3	0.582	x	142.2
4	0.247	y	4.7
5	0.222	Y	10.9
6	0.180	x	8.2
7	0.136	x	1.3
8	0.122	y	3.9
9	0.099	x	2.8
10	0.099	x	0.9
11	0.088	y	2.1
12	0.081	x	0.3

provides detailed results of the modal analysis.

Figure 18 depicts the horizontal displacements brought about by earthquake forces acting along the positive x- and y-axes. As shown in

Table 9. Controls for relative floor displacement (Ra = 1).

Seismic Level	Story	Displacement ∆D, (mm)		Drift Ratio, ∆D/H		Limit Value	Result
		x	y	%	%	%	√/×
SL-3	basement	1.49	2.51	0.05	0.08	0.3 (LD)	√
	ground	2.30	4.08	0.08	0.14	0.3 (LD)	√
	first	2.69	3.40	0.09	0.11	0.3 (LD)	√
	second	2.50	2.70	0.08	0.09	0.3 (LD)	√
	third	2.62	2.13	0.08	0.07	0.3 (LD)	√
	fourth	2.28	1.25	0.07	0.04	0.3 (LD)	√
	fifth	2.75	0.15	0.11	0.06	0.3 (LD)	√
	sixth	3.13	3.60	0.12	0.14	0.3 (LD)	√
SL-2	basement	4.47	5.49	0.15	0.18	0.7 (CD)	√
	ground	7.42	8.59	0.25	0.28	0.7 (CD)	√
	first	8.65	7.86	0.29	0.26	0.7 (CD)	√
	second	7.57	6.97	0.25	0.23	0.7 (CD)	√
	third	7.43	6.58	0.24	0.22	0.7 (CD)	√
	fourth	6.23	5.21	0.21	0.17	0.7 (CD)	√
	fifth	4.62	3.18	0.18	0.13	0.7 (CD)	√
	sixth	4.45	3.23	0.17	0.13	0.7 (CD)	√
SL-1	basement	5.58	9.94	0.19	0.33	1.0 (PC)	√
	ground	9.31	17.17	0.31	0.57	1.0 (PC)	√
	first	10.87	14.90	0.36	0.49	1.0 (PC)	√
	second	9.46	13.68	0.31	0.46	1.0 (PC)	√
	third	9.24	13.39	0.30	0.45	1.0 (PC)	√
	fourth	7.70	11.20	0.26	0.37	1.0 (PC)	√
	fifth	5.70	7.61	0.22	0.30	1.0 (PC)	√
	sixth	5.75	4.56	0.23	0.18	1.0 (PC)	√

, the relative story drifts of the building were computed separately for each of the three earthquake levels and compared to the regulatory limit values. Based on these comparisons, the structure satisfies the performance criteria for “limited damage” under the SL-3 earthquake level, “controlled damage” under the SL-2 earthquake level, and “pre-collapse” under the SL-1 earthquake level.

The masonry structure’s crucial wall sections surrounding the window margins were examined at seismic levels SL-3, SL-2, and SL-1. Figure 19 presents the shear forces in these critical areas, while

Table 10. The performance analysis’s findings.

Seismic Level	Ra	Story	Shear Force	* %r	Target Performance Level	Check
SL-3	1	basement	×	57.53	LD	×
		ground	×	55.09		×
		first	×	67.61		×
		second	×	64.54		×
		third	×	55.32		×
		fourth	×	58.12		×
		fifth	×	55.80		×
		sixth	×	48.24		×
SL-2	1	basement	×	30.92	CD	√
		ground	×	29.79		√
		first	×	58.38		×
		second	×	59.15		×
		third	×	53.04		×
		fourth	×	55.17		×
		fifth	×	48.67		×
		sixth	×	30.28		√
SL-1	3	basement	×	66.82	PC	×
		ground	×	59.43		×
		first	×	68.99		×
		second	×	63.11		×
		third	×	62.84		×
		fourth	×	65.74		×
		fifth	×	61.43		×
		sixth	×	47.88		×

* The ratio of total shear force to shear force acting on the floor appears insufficient to sustain the forces generated by earthquake loading to withstand the forces produced by earthquake loads.

verifies the shear force for three earthquake scenarios. The results indicate that the structure does not meet the “limited damage” performance level for the SL-3 earthquake, the “controlled damage” level for the SL-2 earthquake, or the “pre-collapse” level for the SL-1 earthquake when compared to the permissible limits. Figure 20 illustrates the out-of-plane moments in the critical sections.

5.2. Nonlinear Analysis

An essential method for assessing the nonlinear static behavior of masonry structures is pushover analysis. Finding probable collapse processes and comprehending how a structure reacts to lateral stresses are the goals of nonlinear static analyses. The relationship between base shear force and horizontal displacement, incorporating plastic dissipation, is determined through nonlinear static analysis [50]. This approach is beneficial for retrofitting, protecting historic buildings, and evaluating seismic performance. At this stage, the FEM model used in linear analysis has been employed, with modifications made to the material properties and loading conditions. Nonlinear material properties are described using bilinear idealization, incorporating integrated plastic material models and corresponding requirements. The Strumas masonry material model is employed alongside the homogenization method created by Lee et al. [51], which effectively characterizes masonry structures through an analogous material approach. This methodology is utilized to represent the material nonlinearity of the walls accurately. The study also takes into account the orthotropic material qualities developed by Pande et al. [52,53], taking into account variables such as brick dimensions (width, height, and length), Young’s modulus, Poisson’s ratio, and the thickness of mortar joints that run horizontally and vertically. These equivalent properties are derived using the strain energy method, assuming perfect bonding between bricks and mortar with continuous mortar joints in both directions. The computed orthotropic parameters inform the construction of the stiffness matrix in the finite element model, facilitating the development of structural linkages and accurate stress–strain assessments [54]. Displacement-controlled analyses using the Newton–Raphson method were conducted separately for the positive x- and y-directions. Each structure node was subjected to a 1 kN load in the direction of the static thrust analysis. Following SRMGHS-17 [35] guidelines, a pushover curve was generated to define performance limits. These guidelines classify performance based on drift ratios, as follows: 0.3% signifies “limited damage”, 0.7% denotes “controlled damage”, and 1% indicates “pre-collapse damage”. The 3D model created with the finite element software Midas Gen v2020 [45] was utilized for the pushover study. Structural displacements under lateral loads were evaluated in positive x- and y-directions. According to SRMGHS-17 [35], structures are deemed to collapse if lateral displacements exceed 1% of their height. In this analysis, loads were applied until a displacement of 300 mm was reached, slightly exceeding the 1% criterion for the 23,500 mm tall structure under study. Figure 13 illustrates control points in the x- and y-directions and presents pre-collapse performance data.

Table 11. Properties of masonry wall materials.

Material	Young’s Modulus (MPa)	Poisson’s Ratio	Tensile Strength (MPa)	Stiffness Reduction Factor
brick	2200	0.25	0.22	0.00001
stone	2800	0.25	0.28	0.00001
bed joint	1250	0.25	0.15	0.00001
head joint	1250	0.25	0.15	0.00001

details the plastic properties of the masonry materials used in the model. Drift ratios of 0.3%, 0.7%, and 1% of the structure height were observed for SL-3, SL-2, and SL-1 earthquake levels, respectively. The areas with the most significant displacement in the x- and y-directions at the top of the structure were designated as control points based on the findings of the response spectrum analysis (Figure 21). This approach provides a comprehensive method for understanding and assessing masonry structures’ seismic behavior and performance, which is critical for retrofitting and preservation efforts.

Figure 22 shows representative wall specimens illustrating the loading patterns applied uniformly in each principal direction during the pushover analyses. Figure 23 presents the structure’s static pushover curve. In this graph, the x-axis represents the lateral displacement, while the y-axis corresponds to the base shear force. The structure’s horizontal load capacities are 11,997.09 kN in the +x-direction and 16,209.01 kN in the +y-direction. The vertical dashed lines show the graph’s SL-3, SL-2, and SL-1 earthquake level limit values. The value of 0.3% indicates limited damage, 0.7% indicates controlled damage, and 1.0% indicates pre-collapse performance level.

Figure 24 shows the horizontal displacements caused by the static thrust forces applied in the positive x- and y-axes.

Table 12. Displacement checks for control points.

Seismic Level	Base Shear (kN)		Performance Point (mm)		Drift Ratio, ∆D/H		Limit Ratio	Result
	x	x	+x	+y	%	%	%	√/×
SL-3	10,995.57	8090.55	81.08	65.43	0.036	0.080	0.3 (LD)	√
SL-2	14,361.01	10,604.19	182.10	159.86	0.087	0.197	0.7 (CD)	√
SL-1	16,209.01	11,997.09	258.21	231.35	0.126	0.312	1.0 (PC)	√

displays the relative displacements of specific control points on each building floor concerning the bottom story. Each earthquake scenario’s displacements are computed independently, and the results are compared to the permitted limits. The comparative results indicate that the structure meets the performance standards for the following levels: “controlled damage” under the SL-2 earthquake simulation, “limited damage” under the SL-3 earthquake scenario, and “pre-collapse” under the SL-1 scenario. Figure 25 and Figure 26 illustrate the principal and shear stresses in the masonry elements for the earthquake levels included in the pushover assessments conducted in the positive x- and y-directions. In the graphical presentation of the analysis results, the reinforced concrete floor slabs were deliberately hidden in the model to provide a clearer visualization of the stress distribution within the masonry walls.

6. Discussion

This study thoroughly evaluates the seismic performance of a conventional masonry building with reinforced concrete slabs by employing a multi-stage inquiry strategy and finite element analysis. The evaluation employs numerical methods, including response spectrum and pushover analyses, to assess the structure under three earthquake scenarios (SL-3, SL-2, and SL-1) defined by Turkish seismic codes. The building achieved the inter-story drift limits for all scenarios, as follows: 0.3% for SL-3, 0.7% for SL-2, and 1.0% for SL-1. The maximum relative drift ratios were 0.14%, 0.28%, and 0.57%, respectively. However, the shear force demands exceeded the permissible threshold of 40% of base shear on multiple floors across all seismic levels, reaching up to 68.99% at SL-1. The nonlinear pushover analysis revealed maximum lateral base shear capacities of 11,997.09 kN (x-direction) and 16,209.01 kN (y-direction), with performance points being reached at 258.21 mm and 231.35 mm displacements, respectively. Despite acceptable displacement behavior, inadequate lateral shear strength indicates that structural retrofitting is necessary.

The flat-jack test results further revealed that the masonry walls on the fourth story possessed noticeably greater in situ stiffness than those at the other levels. Such inter-story variability is attributed to the traditional construction practices—including on-site mortar mixing, the absence of standardized brick production during the 1950s, and artistry inconsistencies—which produced heterogeneous material properties. Therefore, the double flat-jack campaign was designed to detect and quantitatively characterize these differences. These findings underscore the vulnerability of traditional hybrid masonry systems and highlight the value of integrating experimental testing and advanced modeling to inform conservation and seismic upgrading strategies.

In modeling masonry structures, the type of floor system critically influences the structural behavior under seismic loads. For buildings with timber floors or precast reinforced concrete slabs of separate units, the connection between floors and masonry walls is typically weak, resulting in limited in-plane stiffness and negligible moment transfer. Therefore, these systems are modeled without rigid diaphragm assumptions, often using pinned or supported connections [3,4]. On the other hand, the floors are assumed to act as rigid diaphragms in structures incorporating monolithic cast-in-place reinforced concrete slabs, such as the one studied in this paper. This configuration enables full moment and shear transfer between the floor system and the surrounding masonry walls, significantly enhancing the lateral load distribution capacity of the structure [29,33]. Consequently, the selection of floor–wall interaction modeling must be tailored to the specific construction type to ensure an accurate representation of seismic performance.

7. Conclusions

This study presented a comprehensive seismic performance evaluation of a traditional masonry building incorporating cast-in-place reinforced concrete slabs based on a multi-stage methodology involving experimental characterization and finite element modeling. The seismic response was assessed under three earthquake scenarios—SL-3, SL-2, and SL-1—defined by Turkish seismic regulations. The in situ material tests revealed that the masonry walls exhibited an elastic modulus ranging from 200.2 MPa to 1062.2 MPa, with an average value of 620.1 MPa and a coefficient of variation of 23.5%. The bed joint shear strength varied between 0.50 MPa and 0.79 MPa, averaging 0.64 MPa, with an 18.2% coefficient of variation, reflecting a moderate level of material heterogeneity characteristic of aging masonry structures. The linear response spectrum analysis demonstrated that the maximum relative inter-story drift ratios remained within the regulatory thresholds, with maximum values of 0.14% for SL-3, 0.28% for SL-2, and 0.57% for SL-1, satisfying the limited damage, controlled damage, and pre-collapse performance levels, respectively. Nevertheless, the shear force demands on critical stories substantially exceeded the allowable 40% of base shear, reaching up to 68.99% at the first floor under SL-1 loading, thereby compromising the building’s overall seismic safety. The nonlinear static pushover analysis indicated that the ultimate base shear capacities were 11,997 kN in the x-direction and 16,209 kN in the y-direction, with corresponding displacement demands of 258.21 mm and 231.35 mm, respectively. Although the structure achieved target drift ratios aligned with the pre-collapse performance level, the insufficient shear strength, particularly at critical wall regions around openings, underscores the necessity for seismic retrofitting interventions. In conclusion, while the building exhibits adequate displacement capacity and global stability under severe seismic excitation, its vulnerability to shear failure necessitates targeted strengthening measures to ensure compliance with modern seismic performance standards. The findings of this study highlight the critical importance of integrating rigorous experimental investigations with advanced numerical analyses for the reliable seismic assessment and preservation of traditional masonry structures. Future research should prioritize the development of retrofit techniques that enhance shear performance without compromising the architectural integrity of historic buildings.