Quantitative Assessment of Seismic Retrofit Strategies for RC School Buildings Using Steel Exoskeletons and Localized Strengthening

Abstract

This study offers a quantitative performance assessment of integrated seismic retrofit designs applied to an in-service 1960s reinforced concrete school structure in Central Italy. The research combines in-depth experimental material characterization with complex numerical simulations in order to estimate both the independent and interaction effects of external steel exoskeletons in conjunction with localized CAM (Cucitura Attiva dei Materiali) strengthening. The experimental investigation includes extensive material characterization through core drilling and non-destructive pacometric inspections to accurately define the existing concrete properties. The numerical analysis is performed with Finite Element modeling to estimate four different structural conditions: the original state, the condition with static strengthening, the condition with additional steel exoskeletons, and the condition with both exoskeletons and localized CAM reinforcements. The results quantitatively estimate the specific performance gains from the individual retrofit strategies. The steel exoskeletons show effective reduction in inter-story drifts but negligible effect on strength-oriented failure mechanisms. Localized CAM strengthening therefore stands out as necessary in reaching adequate safety levels in all the failure mechanisms. Economic analysis reveals that while steel exoskeletons provide the major cost component, the integrated approach with localized strengthening is essential for achieving comprehensive seismic safety enhancement.

1. Introduction

The seismic vulnerability of educational buildings in Europe represents a critical infrastructure challenge for the 21st century. In Italy in particular, almost 55% of the national territory was classified as non-seismic until 1984 and accommodates about 40% of the population in buildings that showed little consideration for seismic design [1]. A large majority of the reinforced concrete (RC) framed schools were constructed in the post-war reconstruction period (through the 1950s–1970s), adopting design practices with overwhelming consideration given to gravity loads and little consideration given to resisting lateral loads [2,3]. Understanding the specific structural vulnerabilities that necessitate integrated retrofit approaches requires examination of the historical construction context and material characteristics of the target building typology.

Statistical analysis of the existing building stock reveals concerning patterns in material quality and structural performance that directly inform the selection of retrofit strategies analyzed in this study. In particular, buildings from the 1962–1971 period have average concrete compressive strengths of only 19.77 MPa, significantly below current design criteria [3]. These material limitations, combined with inadequate reinforcement detailing typical of gravity load design, create the specific failure mechanisms that require both global stiffening and local strengthening addressed in this research. The EU-CENTRE “Progetto Scuole” project, launched in response to the destructive collapse in the 2002 Molise earthquake of the Iovene primary school building, has created a large database covering more than 49,000 schools in Italy, which has recognized three main at-risk categories: reinforced concrete frame buildings with masonry infilling, unreinforced masonry buildings, and prefabricated concrete buildings [2].

Recent seismic events have provided unprecedented empirical data on school building performance under actual earthquake conditions. The L’Aquila earthquake (2009, Mw 6.3) demonstrated significant vulnerabilities even in supposedly modern structures, with the University of L’Aquila dormitory collapse resulting in 55 fatalities [4]. The Emilia-Romagna earthquake sequence (2012) particularly affected precast concrete structures, with approximately 500 industrial facilities severely damaged due to inadequate beam–column connections [5]. Most recently, the Central Italy earthquake sequence (2016–2017) provided detailed instrumental data from monitored school buildings, where peak ground accelerations reached 0.65 g and induced inter-story drifts of 3% in masonry portions exceeding life safety limits by factors of 5–7 [6].

School buildings serve a dual function in an infrastructural system. Beyond basic education requirements, schools qualify as “strategic buildings for the purpose of civil protection” in Italian seismic codes as rescue shelters and incident coordination centers in the event of earthquake recovery efforts [7]. This dual function places those buildings under strict performance needs where the design requirements are based on 712-year return periods as opposed to the 475-year periods in the normal constructions [7].

The retrofit of existing school buildings presents unique technical and operational challenges that distinguish educational facilities from other building typologies [8]. Unlike residential or commercial buildings, schools have the permanent presence of students during the learning year, making traditional invasive retrofitting methods challenging to adopt [9]. The need to keep educational activity despite performing extensive structural alterations has triggered the development of non-invasive retrofitting technologies that can upgrade schools on the exterior side of the building and hence limit interference in regular operations [10].

Economic considerations further contribute to the complexity of strategies in retrofit implementation. Recent studies indicate that there are 6 million buildings in Italy in need of seismic upgrade, where 80% of the existing building stock fails to meet the current seismic standards [11]. The cumulative economic damage from the earthquakes in the years 1944–2017 totals EUR 212 billion, highlighting the need for cost-friendly retrofit measures amenable to large-scale implementation [11].

The use of steel exoskeletons is recognized as a promising retrofitting approach for the significant increase in structural performance they provide [12]. Steel exoskeletons can be defined as external self-supporting structural systems connected to the existing building structure for improving strength, stiffness, and dissipation capacity under seismic actions while contemporarily defining a new building envelope with different functions such as ventilated walls, solar greenhouses, shields, and support for insulating panels or solar panels [13]. A comprehensive classification system for exoskeletons has been developed based on structural material (steel, reinforced concrete, masonry), geometrical features (three-dimensional or planar), placement with respect to the existing structure (orthogonal or parallel to the facade), and seismic-resistant system configuration. Therefore, a planar steel exoskeleton placed perpendicularly to the facade of the existing building with concentric X-shaped bracing could be identified as EXO-S-2D-⊥-CBF-X [14].

Recent research has demonstrated that connection characteristics significantly influence exoskeleton effectiveness, with rigid connections maximizing force transfer, but require careful detailing to accommodate thermal movements and construction tolerances [14,15]. The connection stiffness and strength directly affect the composite behavior between existing and new structural systems, influencing both global response characteristics and local stress concentrations at interface locations [16,17]. Advanced connection strategies employing friction dampers or yielding elements can provide controlled energy dissipation while maintaining structural integrity under seismic loading [18,19,20]. Experimental investigations have documented up to 70% increases in lateral strength capacity and 300% improvements in energy dissipation through optimized connection designs, with connection stiffness matching 50–80% of host structure stiffness proving optimal for composite behavior [17].

Exoskeleton applications have been categorized into non-dissipative and dissipative configurations based on their energy dissipation characteristics. Non-dissipative systems rely primarily on elastic strength and stiffness enhancement through rigid connection systems, diagrid arrangements, and cold-formed steel frameworks, while dissipative systems incorporate energy dissipation devices or connections [16]. These energy dissipation approaches can be implemented within various structural configurations, with diagrid configurations accommodating both non-dissipative and dissipative approaches through appropriate connection and device integration [21].

Recent performance assessments have documented substantial improvements achievable through exoskeleton adoption. Documented case studies report strength increases of 18.5–27.5% in ultimate shear capacity, 8.3% enhancement in initial stiffness, and 9.3–29.5% improvements in ultimate displacement capacity [22]. The seismic safety coefficient can improve from 0.26 to 0.42 in well-documented applications, representing substantial improvement in overall building performance [23]. Steel exoskeletons for integrated seismic and energy retrofit have demonstrated significant improvements in thermal characteristics of building envelopes with relevant reductions in primary energy while preserving elastic behavior of existing structures [13].

Many applications of orthogonal steel exoskeletons for global reinforcement of existing structures have been implemented in Italy for seismic strengthening of schools and industrial buildings. The office building of Magneti Marelli factory in Crevalcore, after structural damage due to the northeastern Italy earthquake in 2012, was retrofitted through X-braced shear walls arranged orthogonally to the facades in both main directions to increase lateral stiffness and resistance [24]. The integrated retrofit of the “La Tina” Primary School in Città di Castello represents another example where structural inadequacies related to poor lateral stiffness and resistance, together with the need to carry out an intervention without interrupting building use, guided designers towards steel exoskeletons [23]. Similar interventions were adopted for the seismic retrofit of the “A. Fiori” school in Formigine and the “M. Montessori” school in San Felice Sul Panaro, where orthogonal exoskeletons were combined with parallel ones due to architectural constraints [14].

The advantage of exoskeleton systems in conjunction with dry and prefabricated modular solutions is the possibility of creating an integrated and reversible retrofit that can adapt to future needs of the building. The double-skin configuration can reduce the exposure class of the existing building while providing the possibility of fully satisfying holistic vision requirements from a life cycle thinking point of view, acting on structural, architectural, and environmental performances [25]. Orthogonal steel exoskeletons are particularly suitable for isolated and low/mid-rise buildings, though geometrical limitations on maximum width at the base of the system determine the adoption of deep foundations necessary to counteract overturning moments and shear at the base [14].

The design optimization of exoskeleton systems involves sophisticated analytical methodologies that have evolved beyond traditional approaches. Advanced computational frameworks utilize multi-objective optimization algorithms to balance structural performance, material efficiency, and construction constraints, enabling designers to achieve optimal configurations for specific building typologies [16]. Dissipative exoskeleton designs require particular attention to the dynamic interaction between energy dissipation devices and the overall structural system, with advanced control strategies ensuring effective seismic response modification while maintaining structural stability [21]. The application of cold-formed steel profiles in exoskeleton construction offers significant advantages in terms of fabrication efficiency and weight reduction, though specialized connection details are required to ensure adequate load transfer and accommodate thermal movements [15].

Beyond seismic performance enhancement, contemporary exoskeleton applications must address additional safety considerations that have emerged from evolving building use patterns. The integration of electric vehicle charging infrastructure in retrofit projects introduces novel fire safety challenges that require careful consideration in exoskeleton design, particularly regarding structural response under extreme thermal loading conditions and evacuation pathway protection [26]. Advanced computational modeling techniques now enable detailed analysis of structural behavior under combined seismic and thermal loading scenarios, providing critical insights for comprehensive safety assessment [27].

Local strengthening approaches recognize the inherent potential within existing structural systems and allow for localized additions to critical elements [28]. As an example, Fiber Reinforced Polymer (FRP) composites have been demonstrated to be effective in applications involving joint and member strengthening [29,30].

Advanced strengthening techniques are designed to address specific failure modes recognized in existing buildings before the adoption of seismic code provisions. Steel-based joint strengthening methods show interesting performance upgrade with steel jacketing increasing deformation and load-carrying capacities by 10.7% and 18.1%, respectively [30]. Moreover, steel haunch insertions provide up to 53.3% upgrade in lateral load-carrying capacity while effectively relocating plastic hinges from vulnerable joint regions to beam members [30].

CAM (Cucitura Attiva dei Materiali) system represents an innovative approach towards localized strengthening by utilizing high-capacity metallic strips to provide active three-dimensional confinement through post-installed closed stirrups [31]. The technique enables a great increase in the structural element capacity by maintaining architectural aesthetic and minimum interference during construction.

Contemporary retrofit strategies increasingly emphasize integrated approaches that combine global and local interventions to optimize both performance and cost-effectiveness. The development of these methodologies benefits from advances in structural dynamics and experimental identification techniques applied to diverse building typologies and complex geometric configurations [32,33]. Integrated seismic and energy interventions could have economic benefits ranging from 20% to 30% compared to non-integrated strategies implemented in sequence [9]. Cross-Laminated Timber (CLT) systems represent environmentally friendly alternatives with favorable mechanical and thermal performance characteristics [34,35,36]. The “nested buildings” approach considers the integration of new inner CLT structures within existing buildings, making this approach particularly suitable for buildings with heritage constraints.

Performance-based evaluation protocols integrate disparate evaluative criteria into unified indicators, such as the Green and Resilient Indicator (GRI), which integrates seismic resilience and energy efficiency in developing holistic protocols for building evaluation [37]. Multi-criteria evaluation protocols enable knowledge-driven decision-making in the implementation of retrofit measures by considering technical performance characteristics, cost constraints, and sustainability targets [38,39].

An accurate assessment of existing building performance always requires a comprehensive material characterization through both destructive and non-destructive testing protocols. In situ material properties often exhibit significant variation from assumed design values, as shown in data analysis of 2010 compression test samples, which revealed a consistent variability across construction periods [3]. Pre-1961 construction shows a mean compressive strength of 28.24 MPa (standard deviation: 14.73 MPa), while 1962–1971 period buildings exhibit reduced mean strength of 19.77 MPa (standard deviation: 9.00 MPa) [3].

The approach in Eurocode 8-3 provides structured methods for the evaluation of uncertainty through knowledge levels (KL1-KL3) and related confidence factors ranging from 1.00 (KL3) through 1.35 (KL1) [40]. Test data demonstrate overestimation by up to 80% in structural capacity in the absence of adequate material characterization and emphasize the need for broad testing efforts [3].

Modern finite element modeling approaches demonstrate high accuracy when properly validated against experimental data. Common numerical tools can provide good modeling capabilities with typical deviations under 12% for load-carrying capacity predictions [41]. Advanced material models incorporate sophisticated constitutive relationships for the confined concrete material, the steel yield strength, and interfacial response in retrofit applications.

Pushover analysis methods have transformed from simple capacity spectrum approaches to complex adaptive methods where the loading model is varied in response to changes in structural response [42]. Performance-based methods involving the use of tools such as SPO2IDA (Static Pushover to Incremental Dynamic Analysis) facilitate the quantification of retrofit efficiency within the range of various performance levels [43].

Despite significant progress in retrofit technologies and evaluation methods, there are still some shortcomings in evaluating whole-building retrofit strategies for schools. The overriding focus in most existing literature is narrowed towards specific retrofit methods, leading to a narrow perspective regarding the integrated methods utilized in learning environment settings.

This study addresses these knowledge gaps through detailed performance assessment of an integrated retrofit strategy applied to a representative RC school building constructed in the 1960s in Central Italy. The main objective is to provide quantitative findings on performance gains from the adoption of steel exoskeleton structural systems and localized strengthening through CAM interventions and clarifying the interaction effects between different retrofit methods in reaching a holistic structural upgrade. The primary innovation lies in the systematic quantification of interaction effects between global steel exoskeleton systems and localized CAM strengthening, providing useful insights for evidence-based retrofit design decisions. While the study employs established retrofit technologies, the integrated assessment methodology and comprehensive documentation of performance interactions represent valuable contributions to retrofit design practice for strategic educational facilities. The research additionally sheds light on the experimental material characterization effect on the vulnerability evaluation accuracy and choices on retrofit design selections, and hence reveals valuable insights into the advantages of adopting strict test protocols in the structural evaluation of existing buildings.

The study focuses specifically on a three-story RC frame school building representative of the vulnerable building typology identified in national risk assessment programs. While results are specific to the analyzed structure and seismic context, the comprehensive documentation provides valuable technical data and methodological insights applicable to the broader challenge of school building retrofit across seismically active regions of Europe.

The research contributes to the technical knowledge base through detailed quantification of retrofit performance improvements, economic analysis of integrated intervention strategies, and validation of assessment methodologies for existing educational facilities. The study does not propose novel retrofit technologies but rather provides well-documented performance data from a representative case study, supporting evidence-based decision-making for school building seismic safety enhancement.

This article is organized into seven main sections. Following this introduction, Section 2 presents the case study building characteristics and the comprehensive experimental material characterization campaign, including pacometric surveys and core drilling procedures. Section 3 describes the numerical modeling approach and baseline seismic assessment of the existing structure using commercial FEM software, named Sismicad 13. In Section 4, preliminary results of the vulnerability assessment are presented. Section 5 details the retrofit design strategy, covering both the steel exoskeleton system configuration and the localized CAM strengthening interventions. Section 6 presents the quantitative performance assessment results, comparing individual and combined retrofit contributions through modal analysis and pushover procedures with a focus on the economic analysis of the integrated retrofit approach, including cost breakdown and implementation considerations. Finally, Section 7 discusses the technical implications, study limitations, and conclusions drawn from this comprehensive case study assessment.

2. Case Study

The case study building is located in Colonnella, Teramo province, in the Abruzzo region of Central Italy. The structure houses a comprehensive educational institute (Istituto Comprensivo Statale) serving multiple municipalities, including Colonnella, Controguerra, and Corropoli. The building accommodates primary school facilities on the ground floor (9 classrooms for 148 students and 23 teachers) and secondary school facilities on the first floor (5 classrooms for 75 students and 14 teachers). Additional spaces include a gymnasium, computer laboratory, multimedia language laboratory, science laboratory, and art classroom distributed across the remaining areas of the ground and second floors.

The educational complex consists of three main structural blocks constructed during different periods, representing typical construction practices of their respective eras. Block 1, the earliest construction, features traditional load-bearing masonry in solid brick with lime mortar, brick-concrete floor systems, and extends over two above-ground levels (approximately 216 m² each) plus a partial basement (approximately 50 m²). Block 3, the most recent addition, employs reinforced concrete frame construction with brick-concrete floor systems across four above-ground levels (approximately 146 m² each). The masterplan of the school with the three-block definition is shown in Figure 1. In

Table 1. Summary comparison of structural blocks in the educational complex.

Block	Construction Period	Structural System	Levels	Floor Area	Study Focus
1	1923	Load-bearing masonry with lime mortar	2 + partial basement	~216 m2 each	Not analyzed
2	1955–1960	RC frame with masonry infill	3	~500 m2 each	Main focus of study
3	2008	Modern RC frame	4	~146 m2 each	Not analyzed

, a summary of the main characteristics of each block is reported.

The focus of this study is Block 2, which represents the most extensive portion of the complex and exhibits structural characteristics typical of 1960s Italian school construction. This block features a reinforced concrete frame structure with brick-concrete floor systems, extending over three above-ground levels of approximately 500 m² each. The structural system consists of a regular grid of reinforced concrete columns and beams, with masonry infill walls providing non-structural enclosure and partition functions. The floor plan of Block 2 is shown in Figure 2.

The building site is located within a seismically active region of Central Italy, characterized by significant tectonic activity along the convergence zone between the Eurasian and African plates. According to the Italian seismic classification system established by OPCM 3274/2003, the Teramo province is classified as Zone 2, corresponding to high seismicity with expected peak ground acceleration (PGA) values between 0.15 g and 0.25 g.

Local seismic hazard assessment was conducted using the Level I seismic microzonation study for the Municipality of Colonnella. Analysis of the Homogeneous Microzone Map for Seismic Perspective indicates that the building site is classified within stable zones susceptible to local amplification effects. Based on geological, geomorphological, and lithostratigraphic observations, the soil profile is classified as Type C according to Eurocode 8 criteria, corresponding to deposits of very dense sand, gravel, or very stiff clay with thickness greater than 30 m and average shear wave velocity (vs30) between 180–360 m/s.

The specific seismic microzone exhibits a stratigraphic profile characterized by alternating layers of clayey and sandy deposits overlying bedrock formations. Geotechnical parameters derived from the regional microzonation study indicate standard penetration test (SPT) N-values ranging from 15–30 in the upper soil layers, increasing with depth to values exceeding 50 near the bedrock interface at approximately 25–30 m depth.

Experimental Material Characterization Campaign

A comprehensive experimental program was implemented to accurately characterize the mechanical properties of existing structural materials. The investigation strategy combined non-destructive and minimally destructive techniques to minimize impact on building operations while obtaining reliable material property data [43,44]. The testing program was designed in accordance with Italian standards for existing building assessment (NTC 2018, Chapter 8 [45]) and international guidelines for in situ concrete evaluation. The floor plan with the element tested is shown in Figure 3.

Non-destructive testing employed pacometric surveys to map reinforcement locations, diameters, and cover depths throughout representative structural elements. These surveys utilized a ZBL-K630A PK-R6 pacometer based on eddy current principles, offering enhanced stability and accuracy compared to conventional electromagnetic devices. The instrument system included a battery-operated control unit with calibration and measurement controls, digital display for direct reading, and directional magnetic coil probes (both small probe and large scanner configurations).

Pacometric investigations were conducted on selected structural elements across both ground and first floor levels to establish representative reinforcement patterns typical of 1960s construction practice. The surveys revealed reinforcement configurations consistent with gravity load design approaches, with minimal consideration of seismic detailing requirements.

Ground floor beam elements exhibited longitudinal reinforcement consisting primarily of 10–18 mm diameter bars with 8 mm stirrups spaced at 150–280 mm centers. Column elements showed 4–16 mm diameter longitudinal bars with 8 mm ties at 160 mm spacing. Concrete cover measurements ranged from 20–35 mm, generally meeting contemporary construction standards but showing significant variability across different elements. In

Table 2. Pacometric survey results for ground floor elements.

Element	Measured Dimensions [cm]	Reinforcement Detected and Estimated on Site (+/− Φ)
Pillar 22	30 × 50	4 Φ 16—stirrups Φ8 every 15 cm
Pillar 10	30 × 30	4 Φ 16—stirrups Φ8 every 18 cm
Pillar 37	30 × 30	4 Φ 16—stirrups Φ8 every 15 cm
Pillar 41	15 × 30	4 Φ 16—stirrups Φ6 + Φ8 every 15 cm
Pillar 28	30 × 30	4 Φ 16—stirrups Φ6 + 10 every 10 cm
Pillar 38	30 × 30	4 Φ 16—stirrups Φ8 every 9 ÷ 15 cm
Pillar 11	30 × 30	4 Φ 16 ÷ 12—stirrups Φ8 every 20 cm
Pillar 26	30 × 30	4 Φ 20—stirrups Φ10 every 15 cm
Pillar 31	30 × 30	4 Φ 16—stirrups Φ8 every 15 cm
Beam 13-6	B = 35 H = slab	2 Φ 16 Φ—stirrups Φ8 every 20 cm
Beam 10-11	B = 42 H = 35 + slab	2 Φ 16 + 7—stirrups Φ8 every 15 ÷ 25 cm
Beam 11-12	B = 100 H = slab	13 Φ 18—stirrups Φ8 every 15 cm
Beam 21-22	B = 110 H = slab	13 Φ 16 + 7—stirrups Φ8 every 15 cm

the results of pacometric for ground floor elements are presented.

First floor investigations revealed similar reinforcement patterns, with slight variations in bar sizes and spacing reflecting construction sequencing. The consistent use of ribbed reinforcement (FeB 44k grade) throughout the structure confirmed adherence to 1960s Italian reinforcement standards, though quantities remained limited compared to modern seismic design requirements. In

Table 3. Pacometric survey results for first floor elements.

Element	Measured dimensions [cm]	Reinforcement detected and estimated on site (+/− Φ)
Pillar 32	44 × 38 + wall	4 Φ 16—stirrups Φ8 every 16 cm
Pillar 25	33 × 33	4 Φ 16—stirrups Φ8 every 15 cm
Pillar 41	16 × 30	4 Φ 16—stirrups Φ8 every 20 cm
Pillar 22	33 × 33	4 Φ 16—stirrups Φ8 every 20 cm
Pillar 12	32 × 32	4 Φ 16—stirrups Φ6 ÷ 8 every 15 cm
Pillar 10	47 × 32	4 Φ 16—stirrups Φ8 every 15 cm
Pillar 26	32 × 32	4 Φ 16—stirrups Φ10 every 15 ÷ 18 cm
Pillar 24	32 × 32	4 Φ 16—stirrups Φ8 every 16 cm
Pillar 47	10 × 30	4 Φ 16—stirrups Φ8 every 16 cm
Pillar 13	30 × 30	4 Φ 14 + 16—stirrups Φ8 every 12 cm
Beam 26–27	B = 30 H = slab	2 Φ 14 + 2 Φ 16—stirrups Φ8 every 17 cm
Beam 10–11	B = 40 H = 35 + slab	2 Φ 16—stirrups Φ10 every 10 cm
Beam 33–41	B = 20 H = 15 + slab	3 Φ 14 + 3 Φ 14—stirrups Φ8 every 20 cm

the results of pacometric for first floor elements are presented.

Important observations included the absence of transverse reinforcement at beam–column joint regions, confirming the gravity load design approach and highlighting potential seismic vulnerabilities. Stirrup configurations showed open rather than closed ties in many locations, further indicating limited consideration of seismic loading requirements during original design and construction.

Destructive testing employed diamond core drilling to extract cylindrical specimens for laboratory compression testing. Core locations were strategically selected based on pacometric survey results to avoid reinforcement and ensure representative sampling of concrete quality throughout the structure. A total of 11 cores were extracted: 5 from column elements and 6 from beam elements across different floor levels.

Core extraction was performed using a COMER electric diamond core drilling machine with 100 mm diameter diamond crown bits. Drilling operations were conducted with continuous water cooling to prevent thermal damage to concrete specimens. Extraction locations were immediately sealed with non-shrink expansive cement mortar to restore structural continuity and prevent moisture ingress.

Core extraction procedures followed UNI EN 12504-1 standard [46] for compressive strength determination, while pacometric surveys employed electromagnetic testing protocols consistent with ASTM C597 guidelines. The LC2 knowledge level classification was selected based on the limited scope of destructive testing (11 cores from 78 total elements requiring assessment) and moderate comprehensiveness of non-destructive investigations, following NTC 2018 criteria [45] for existing building assessment. The corresponding confidence factor of 1.2 accounts for uncertainty in material properties and structural detailing typical of limited knowledge conditions, while avoiding overly conservative assumptions that could lead to unnecessary intervention costs.

Laboratory testing followed EN 12390-3 [47] procedures for compressive strength determination. Specimens were capped with sulfur mortar to ensure parallel loading surfaces and tested using a calibrated compression testing machine at a loading rate of 0.6 ± 0.4 MPa/s. The core extraction and testing program results are summarized in

Table 4. Core specimens extracted from structural elements; d is the diameter of the cylindrical specimen after cutting, H is the height of the cylindrical specimen after cutting, and A_c is the cross-sectional area of the specimen (mm²). H/d represents the height-to-diameter ratio, while k is the correction coefficient applied to cylindrical specimens. f_car indicates the breaking load resistance (N/mm²), and R_c,is represents the estimated cubic resistance in situ.

Specimen	d [mm]	H [mm]	Ac [mm2]	H/d	k	fcar [N/mm2]	Rc, is	Element Reference
C1 COL	74	74	4301	1.000	0.887	28.5	30.3	P10
C2 COL	74	74	4301	1.000	0.887	25.3	26.9	T5-10
C3 COL	74	74	4301	1.000	0.904	19.4	21.0	T6-13
C4 COL	74	74	4301	1.000	0.822	42.9	42.3	P31
C5 COL	74	74	4301	1.000	0.871	34.9	36.5	P25
C6 COL	74	74	4301	1.000	0.822	40.1	39.5	P22
C7 COL	74	74	4301	1.000	0.887	29.9	31.8	T12-5
C8 COL	74	74	4301	1.000	0.887	28.6	30.5	T10-11
C9 COL	74	74	4301	1.000	0.822	41.5	40.9	P7
C10 COL	74	75	4301	1.014	0.892	25.6	27.4	T13-6
C11 COL	74	75	4301	1.014	0.900	23.8	26.6	T12-5

and

Table 5. Summary of Core Compression Test Results.

Element Type	Number of Cores	Mean Strength [MPa]	Standard Deviation [MPa]	Characteristic Strength [MPa]	Concrete Class
Columns Beams	5	37.58	5.91	28.24	C24/28
	6	25.35	3.77	19.17	C20/24

, providing the material property data essential for numerical modeling.

Compression testing results revealed significant variability in concrete strength, both between different elements and within individual structural components. Table 1 summarizes the statistical analysis of core compression test results for beam and column elements separately, reflecting the observed systematic differences between these element types.

The systematic difference between beam and column concrete strengths suggests potential construction sequencing effects or quality control variations during the original construction period. Column elements consistently exhibited higher compressive strengths, possibly reflecting different concrete mix designs or improved consolidation practices for vertical elements.

Statistical analysis employed a confidence level of 95% with appropriate reduction factors for small sample sizes according to fib Model Code procedures. The characteristic concrete strength (f_ck) was calculated as the mean strength minus 1.64 times the standard deviation, corresponding to a 5% fractile value for normal distribution assumptions.

Based on the comprehensive investigation program, the structural assessment knowledge level was classified as LC2 (Limited Knowledge) according to NTC 2018 criteria [45]. The corresponding confidence factor (FC) of 1.2 was applied to reduce material strength values for safety assessment purposes. This factor accounts for uncertainty in material properties and structural detailing typical of limited knowledge conditions.

Material properties for numerical modeling were established as follows:

• Column concrete: f_ck = 27.89 MPa, E_cm = 26,583 MPa (reduced by FC = 1.2);
• Beam concrete: f_ck = 19.17 MPa, E_cm = 23,621 MPa (reduced by FC = 1.2);
• Reinforcement: FeB 44k, f_yk = 430 MPa (typical for 1960s construction).

The material characterization results revealed concrete strengths significantly below modern design expectations, confirming the need for comprehensive structural strengthening. The systematic difference between beam and column concrete properties required element-specific material assignments in numerical models, improving assessment accuracy compared to uniform material assumptions.

The identified reinforcement deficiencies, including inadequate transverse reinforcement and poor joint detailing, confirmed the vulnerability of the existing structure to seismic loading. These findings provided interesting input for designing targeted strengthening interventions, with particular attention to joint regions and inadequately confined elements.

The comprehensive experimental program yielded confidence factors appropriate for detailed retrofit design, avoiding overly conservative assumptions that could lead to unnecessary intervention costs while ensuring adequate safety margins for the strengthened structure.

3. Finite Element Model Development

Numerical analysis was performed using Sismicad, a commercial finite element software specifically designed for seismic assessment of existing buildings according to Italian seismic codes (NTC 2018 [45]). The software platform was selected for its widespread use in professional practice, comprehensive implementation of Italian seismic standards, and specific capabilities for existing building assessment, including knowledge level considerations and confidence factor applications.

The structural model was developed as a three-dimensional frame system with beam and column elements represented using fiber-section formulations to capture nonlinear material behavior. Floor systems were modeled using rigid diaphragm assumptions, appropriate for the brick-concrete floor construction typical of 1960s Italian buildings. Foundation conditions were represented through fixed base constraints, consistent with the study’s focus on superstructure retrofit effectiveness.

3.1. Geometric Modeling and Element Properties

The structural model was developed as a three-dimensional frame system representing the complete geometry of Block 2. The finite element model accurately captured all primary frame elements, including beams and columns, with floor systems modeled using rigid diaphragm assumptions appropriate for the brick-concrete floor construction typical of 1960s Italian buildings. Model geometry was established from architectural and structural drawings and verified through field measurements during the investigation campaign. A representation of the model in Sismicad 13 is reported in Figure 4.

The structural system consists of a regular grid of reinforced concrete elements with varying cross-sectional dimensions determined from original construction drawings and field verification. Column elements were modeled with square cross-sections of either 300 × 300 mm or 400 × 400 mm, depending on their structural function and loading requirements. Beam elements exhibited greater dimensional variety, reflecting the diverse structural roles within the frame system. Deep beam sections included dimensions of 400 × 800 mm, 300 × 700 mm, and 300 × 500 mm, while flush beam configurations employed 800 × 350 mm, 1000 × 250 mm, and 800 × 250 mm sections. Floor systems were represented through 350 mm thick brick-concrete slabs providing both structural continuity and mass distribution for dynamic analysis.

Mass distribution was calculated, incorporating dead loads from structural and non-structural elements along with appropriate fractions of live loads according to NTC 2018 [45] load combination requirements. Dead load components included structural self-weight, masonry infill panels, floor finishes, and building services. Live load participation was considered at 30% for school occupancy classification, reflecting the intermittent nature of educational facility loading patterns. Masonry infill panels were considered as non-structural elements in the analysis, contributing to the overall mass distribution for dynamic analysis, without providing additional stiffness or strength contributions to the structural system. While masonry infill panels are modeled as non-structural elements contributing only to mass distribution, the literature demonstrates that infills can significantly influence seismic response through stiffness and strength contributions, even when not designed as part of the lateral force resisting system. However, several factors support the adopted conservative approach. Current Italian seismic codes (NTC 2018 [45]) do not require explicit inclusion of infill stiffness in analysis for existing buildings, allowing engineers to neglect this contribution. The irregular distribution and variable quality of infill panels in 1960s construction create uncertainty in stiffness quantification, while infill panels are subject to brittle failure mechanisms and strength degradation under cyclic loading, making their contribution unreliable for seismic design. Furthermore, the adopted approach provides results favoring safety by potentially underestimating structural stiffness and overestimating displacement demands.

3.2. Material Property Assignment

Material properties were assigned based on experimental characterization results described in Section 2, incorporating confidence factor reductions as required for LC2 knowledge level assessment. The systematic difference between beam and column concrete strengths identified through testing was explicitly represented in the numerical model through element-specific material assignments, ensuring accurate representation of actual structural capacity rather than idealized uniform properties.

Table 6. Steel properties in FE model.

Description	E [daN/cm2]	ν	γ [daN/cm3]	α [°C−1]	fyk [daN/cm2]	σamm [daN/cm2]	Bar Type	Knowledge Level
FeB 44k improved adherence LC2	2,060,000	0.3	0.00785	0.000012	4300	2350	Improved adherence	LC2 (FC = 1.2)

and

Table 7. Concretes properties in FE model.

Description	E [daN/cm2]	G [daN/cm2]	ν	γ [daN/cm3]	α [°C−1]	Rck [daN/cm2]	Max Aggregate Diameter [cm]	Knowledge Level
C24/28 LC2	265,827.99	120,830.9	0.1	0.0025	0.00001	285	1.5	LC2 (FC = 1.2)
C20/24 LC2	236,214.78	107,370.35	0.1	0.0025	0.00001	239	1.5	LC2 (FC = 1.2)

summarize the main properties of the materials, as defined in the numerical environment.

Constitutive models employed standard material relationships implemented in Sismicad, including concrete behavior with tension cutoff and compression softening characteristics. Steel reinforcement was modeled using bilinear elastoplastic behavior with kinematic hardening appropriate for cyclic loading conditions. Ultimate strain limits were established at 10% for steel reinforcement, reflecting typical values for 1960s Italian construction practice.

Reinforcement configurations were modeled based on pacometric survey results and representative detailing typical of 1960s construction practice. The experimental investigation provided specific reinforcement data for surveyed elements, while remaining elements were assigned reinforcement patterns consistent with contemporary design approaches and construction standards. Column elements typically incorporated 4 Φ 16 mm longitudinal bars with Φ 8/160 mm ties, while beam longitudinal reinforcement varied from 6 Φ 18 mm to 16 Φ 18 mm depending on span and loading requirements. Beam transverse reinforcement consisted of Φ 8 stirrups at 150–280 mm spacing. The absence of adequate joint reinforcement and modern seismic detailing was explicitly represented in the model, ensuring accurate capture of expected failure mechanisms under seismic loading.

Joint behavior was assessed through capacity verification procedures following NTC 2018 [45] provisions for beam–column connections. The joint region capacity was evaluated based on the concrete strength and the available transverse reinforcement within the joint core, which was minimal in the existing 1960s construction due to gravity load design practices. Joint shear strength was calculated according to code formulations considering the effective joint area, concrete compressive strength, and confinement provided by transverse reinforcement.

The baseline joint capacity assessment revealed severe deficiencies typical of pre-seismic code construction, where joints lack adequate transverse reinforcement and rely primarily on concrete shear strength. The CAM system’s active confinement was implemented in the analysis through enhanced material properties reflecting the three-dimensional confining pressure generated by pre-stressed metallic strips. The enhanced concrete properties were calculated using established confinement models (Mander et al.), where the confining pressure is determined from CAM strip tension (controlled through pre-stressing procedures) and the geometric configuration of the strip layout.

The active nature of CAM confinement provides immediate enhancement independent of member deformation, unlike passive strengthening systems. The CAM strips create continuous confining pressure around the joint core region, effectively transforming the unconfined concrete to confined concrete with enhanced compressive strength and ductility. The significant improvement in joint capacity (from multiplier of 0.037 to above unity) reflects the transition from essentially unconfined joint cores typical of 1960s construction to adequately confined joints meeting modern seismic requirements through enhanced shear transfer capacity and improved ductility characteristics.

3.3. Boundary Conditions

Foundation conditions were represented through fixed base constraints, consistent with the study’s focus on superstructure retrofit effectiveness rather than soil–structure interaction effects. This assumption was considered appropriate given the relatively stiff soil conditions (Type C) identified through geotechnical investigation and the primary interest in evaluating retrofit intervention performance.

Beam and column elements were connected through rigid joint assumptions reflecting monolithic concrete construction. Floor diaphragms were modeled as rigid bodies within their own planes while allowing out-of-plane flexibility, appropriately representing the structural behavior of brick-concrete floor systems with adequate in-plane stiffness for force distribution.

The independent exoskeleton foundations were designed with about 500 mm separation distances from existing foundations to prevent interaction effects. The school building site gave enough construction space around the building perimeter for exoskeleton installation without interference from adjacent structures.

3.4. Analysis Procedures

Initial static analysis verified structural adequacy under gravity loads alone, confirming that the existing structure could support current loading conditions without immediate safety concerns. Load combinations followed NTC 2018 [45] requirements for Ultimate Limit State verification. Static analysis results confirmed adequate capacity for gravity load resistance, with utilization ratios remaining below unity for all structural elements.

Dynamic characteristics were established through modal analysis to identify fundamental periods, mode shapes, and participating mass ratios [48,49]. Modal analysis considered sufficient modes to achieve 85% cumulative participating mass in both horizontal directions, as required by NTC 2018 [45]. Advanced dynamic identification techniques have been successfully applied to various structural systems, including high-rise buildings and flexible roof structures, demonstrating the effectiveness of experimental modal analysis for structural characterization [50,51,52].

The first three modes captured the primary dynamic response characteristics, with Mode 1 exhibiting translational behavior in the X-direction at a period of 0.52 s and 65% participating mass, Mode 2 showing translational response in the Y-direction at 0.48 s with 68% participating mass, and Mode 3 demonstrating torsional behavior about the vertical axis at 0.42 s with 12% participating mass. The relatively short fundamental periods reflected the stiff structural system typical of low-rise reinforced concrete frame construction.

Pushover analysis was performed to evaluate structural capacity and identify the main failure mechanisms. Control node displacement was monitored at the center of mass of the top floor level, with capacity curves developed to assess structural performance under increasing lateral loads.

4. Seismic Vulnerability Assessment

The baseline assessment revealed significant seismic vulnerabilities consistent with pre-seismic code construction practices. Multiple failure mechanisms were identified across different limit states, indicating inadequate capacity for modern seismic performance requirements. Specific vulnerabilities included insufficient shear capacity in many beam and column elements, inadequate joint strength, poor ductility resulting from limited transverse reinforcement, and excessive inter-story displacement under design seismic loads. The results of the elements safety check are shown in Figure 5.

Quantitative assessment employed capacity/demand ratios for multiple performance criteria according to NTC 2018 [45] verification procedures. The baseline performance results demonstrated severely inadequate structural response across all failure mechanisms. Shear failure exhibited a multiplier of 0.074 corresponding to a 3-year return period and 0.027 g PGA. Flexural failure showed a multiplier of 0.084, equivalent to a 4-year return period and 0.031 g PGA. Joint failure represented the most important deficiency with a multiplier of only 0.037, corresponding to a 1-year return period and 0.016 g PGA. Inter-story drift achieved a performance ratio of 0.758, corresponding to a 27-year return period and 0.073 g PGA.

The baseline assessment revealed that beam–column joint failure represents the most critical vulnerability, with joint shear capacity insufficient to resist even minimal seismic demands (multiplier 0.037). This deficiency is characteristic of 1960s construction, where joints relied primarily on concrete shear strength without adequate transverse reinforcement. The absence of closed stirrups within joint regions, combined with limited joint core confinement, creates severe vulnerabilities under seismic loading conditions. The quantitative assessment results are summarized in

Table 8. Baseline Seismic Performance Assessment Results.

Failure Mechanism	TR	PGA	IR,TR	IR,PGA
Exceeding chord rotation limit for R.C. columns [SLD]	9	0.044	0.419	0.376
Exceeding chord rotation limit for R.C. columns [ULS]	197	0.181	0.59	0.628
Shear failure of R.C. columns [ULS]	155	0.163	0.535	0.565
R.C. column joint failure [ULS]	0	0	0	0
Exceeding chord rotation limit for R.C. beams [SLD]	0	0	0	0
Exceeding chord rotation limit for R.C. beams [ULS]	75	0.117	0.397	0.407
Exceeding shear limit for R.C. beams [ULS]	0	0	0	0
Exceeding inter-story displacement limit [SLO] C7.8.1.5.4	7	0.039	0.466	0.424
Exceeding inter-story displacement limit [SLD] C7.8.1.5.4	9	0.044	0.419	0.376

, which presents the baseline performance multipliers across all critical failure mechanisms.

Seismic risk classification was determined according to Italian standards for existing buildings, considering both economic losses (PAM—Perdita Annuale Media) and safety index (IS-V). The baseline structure was classified as Risk Class G, representing the lowest classification and indicating very high seismic risk requiring immediate attention. This classification incorporated both PAM assessments showing greater than 8.5% expected annual loss and IS-V evaluation indicating safety index values below 0.1. The overall Risk Class G confirmed the urgent need for comprehensive seismic strengthening to achieve acceptable safety levels for continued building occupancy.

Detailed element-level assessment identified specific structural components requiring priority attention in retrofit design. A total of 78 beam elements failed to satisfy strength requirements, while all column elements exhibited inadequate capacity for combined axial and lateral loading. Beam–column joint regions consistently failed verification due to absence of adequate transverse reinforcement and poor confinement.

The comprehensive baseline assessment confirmed the need for both global and local strengthening interventions. The identification of multiple concurrent failure mechanisms indicated that effective retrofit would require integrated approaches addressing capacity deficiencies across all performance criteria.

5. Retrofit Design Strategy

The comprehensive vulnerability assessment described in Section 3 revealed multiple concurrent failure mechanisms requiring an integrated retrofit approach. The systematic deficiencies identified across shear capacity, joint strength, and ductility characteristics indicated that single-intervention strategies would be insufficient to achieve adequate safety levels. Consequently, the retrofit design employed a dual-strategy approach combining global structural enhancement through external steel exoskeletons with targeted local strengthening using CAM (Cucitura Attiva dei Materiali) systems.

The selection of this integrated approach was driven by both technical performance requirements and practical implementation considerations. Steel exoskeletons offer the advantage of external application without disrupting building occupancy, while providing global stiffness enhancement to address inter-story drift concerns. Localized CAM strengthening addresses specific capacity deficiencies in critical structural elements, particularly targeting the joint vulnerabilities and shear capacity limitations identified in the baseline assessment.

5.1. Steel Exoskeleton System Design

The steel exoskeleton system was designed as external structural frames positioned perpendicularly to the building facades, connected to the existing reinforced concrete frame at each beam–column joint location. The exoskeleton configuration was constrained by site conditions, particularly the adjacency to other structural blocks within the educational complex that limited placement options along certain building perimeters.

The exoskeleton system was designed to resist the full seismic design forces according to NTC 2018 [45] requirements for strategic buildings. The system provides multiple beneficial effects: (1) increased global stiffness reducing displacement demands, (2) load sharing that reduces demands on existing structural components, and (3) enhanced overall building strength through the composite structural response. The placement of the exoskeleton systems is schematically shown in Figure 6.

A total of 23 steel exoskeleton units were positioned around the building perimeter, with 9 units along the X-direction and 14 units along the Y-direction. Each perimeter column was connected to at least one exoskeleton unit, with 7 corner columns accommodating connections to two exoskeleton units due to their favorable geometric position. The exoskeleton structures extend over the full three-story height of the building, providing continuous lateral support across all floor levels. The representation of the model with the exoskeleton is also shown in Figure 7.

Each exoskeleton unit consists of vertical steel columns, horizontal beams at each floor level, and diagonal bracing elements configured in X-pattern arrangements. The bracing systems provide in-plane stiffness and stability while transferring lateral forces from the existing structure to the exoskeleton foundation system. Independent foundation systems were designed for each exoskeleton unit, isolated from the existing building foundations to avoid potential interaction effects during seismic loading.

The steel exoskeleton elements were designed using S275 grade steel with non-dissipative behavior characteristics. Initial analysis required iterative design to ensure all steel elements satisfied strength and stability requirements under combined gravity and seismic loading conditions. The iterative process led to the adoption of HEA360 for columns, HEA 300 for beams and HEA 180 for bracing.

Connections between the steel exoskeletons and existing reinforced concrete structure were designed as rigid links positioned at beam–column joint locations. These connections transfer both axial and flexural forces from the existing frame to the exoskeleton system, creating a composite structural response under lateral loading conditions. The connection design employed steel plates welded to exoskeleton members and anchored to the existing concrete frame through post-installed mechanical anchors.

Exoskeleton-to-structure connections employ M20 Grade 8.8 post-installed anchors with 150 mm embedment depth, positioned at 600 mm spacing around the joint perimeter. Steel connection plates (400 × 300 × 15 mm) are welded to exoskeleton members and secured through 4-bolt patterns to distribute concentrated loads. Connection capacity was designed for 150 kN ultimate tension and 200 kN ultimate shear per connection point, with load transfer occurring primarily through direct bearing and anchor shear resistance.

This connection strategy maximizes force transfer from the existing frame to the exoskeleton system, creating composite structural response under lateral loading. The rigid connection assumption represents a conservative approach that ensures effective load sharing while providing upper-bound estimates of exoskeleton contribution to structural response.

The assumption of rigid links represents a simplification that maximizes force transfer to the exoskeleton system, for safety reasons in the estimation of the response. Future studies could investigate the actual connection stiffness characteristics and their influence on system performance. Connection locations were strategically positioned to minimize disruption to existing architectural elements while ensuring effective force transfer. Each connection point was detailed to accommodate the dimensional tolerances typical of existing construction while providing reliable structural continuity between the new and existing structural systems.

The effectiveness of exoskeleton systems requires adequate floor system capacity to distribute forces between the existing frame and external steel elements. The brick-concrete floor systems typical of 1960s Italian construction generally provide adequate in-plane stiffness for force distribution, though detailed verification of floor capacity was not explicitly conducted in this study. The rigid diaphragm assumption employed in the analysis presumes adequate floor capacity for force transfer, which represents a limitation that should be verified in detailed design applications. The assumption of adequate floor capacity is generally reasonable for brick-concrete floor systems of the era, which typically exhibit substantial in-plane stiffness due to the composite action between brick elements and concrete topping. However, this assumption should be verified through detailed analysis of floor system capacity to resist the additional forces introduced by exoskeleton systems, particularly at connection points where concentrated loads are transferred between structural systems.

5.2. CAM Strengthening System

The CAM (Cucitura Attiva dei Materiali) system represents an innovative approach to localized structural strengthening through active confinement of existing reinforced concrete elements. The technology employs high-performance metallic strips arranged as closed loops around structural members, creating three-dimensional confinement effects that enhance both strength and ductility characteristics. The system functions as post-installed transverse reinforcement, addressing the inadequate stirrup provisions typical of pre-seismic code construction.

CAM installations create active tension in the metallic strips through controlled pre-stressing procedures, generating immediate confinement effects independent of member deformation. This active confinement approach provides enhanced performance compared to passive strengthening systems that require significant deformations to activate their contribution. The metallic strips are positioned using corner angles that distribute confining stresses while protecting the strips from direct concrete contact.

All column elements in the existing structure required strengthening based on the baseline vulnerability assessment. The CAM strengthening approach for columns involved increasing the concrete cross-section from 400 × 400 mm to 500 × 500 mm while incorporating the CAM reinforcement system within the enlarged section. This approach provided both additional concrete volume and enhanced confinement through the integrated CAM system. As an example, the detail of the CAM strengthening system on one columns is shown in Figure 8.

The column strengthening procedure employed standard CAM metallic strips with 200 × 20 mm cross-sectional dimensions positioned at regular intervals along the column height. Corner angles provided local stress distribution while maintaining the required spacing for effective confinement. The enlarged concrete sections were cast using high-performance concrete to ensure compatibility with existing materials while providing enhanced mechanical properties.

A total of 78 beam elements failed to satisfy strength requirements in the baseline assessment, necessitating targeted strengthening interventions. CAM strengthening for beams employed external application of metallic strips around the beam perimeter, with concrete section enlargement of 50 mm on each side to accommodate the CAM installation and provide enhanced structural capacity. As an example, in Figure 9, is also shown the CAM reinforcement applied to one beam element.

The beam strengthening approach maintained existing architectural clearances while providing significant capacity enhancement. CAM strips were positioned around the full beam perimeter using appropriate corner protection and spacing requirements. The additional concrete cover provided both protection for the CAM system and enhanced flexural capacity through increased effective depth.

Beam–column joint regions represented the most relevant vulnerability identified in the baseline assessment, with essentially all joints failing to satisfy modern seismic requirements. Joint strengthening employed specialized CAM configurations designed to enhance shear transfer capacity and confinement within the joint core region. The joint CAM applications used larger metallic strips (dimensions adjusted for joint requirements) while maintaining compatibility with the beam and column strengthening systems.

Joint strengthening procedures require careful coordination with adjacent beam and column CAM installations to ensure structural continuity and effective force transfer. The joint CAM systems were designed to address the absence of adequate transverse reinforcement typical of gravity load designed joints while providing enhanced ductility and energy dissipation capacity.

6. Results and Performance Assessment

The performance assessment of retrofit interventions employed the same analytical framework established for the baseline evaluation, ensuring consistent comparison of structural response across different configurations. Four distinct structural models were analyzed to quantify the individual and combined contributions of each retrofit strategy: the baseline existing structure, the statically reinforced structure with enhanced reinforcement, the structure with added steel exoskeletons, and the fully integrated system combining exoskeletons with localized CAM strengthening.

Each model was subjected to identical loading conditions and analysis procedures including static analysis for gravity loads, modal analysis for dynamic characteristics, and pushover analysis for seismic capacity assessment. The systematic comparison approach enabled quantification of performance improvements attributable to specific retrofit interventions while identifying the necessity for integrated approaches to achieve comprehensive safety enhancement.

6.1. Steel Exoskeleton Performance

The addition of steel exoskeletons to the statically reinforced structure produced significant modifications in structural response characteristics, particularly affecting global behavioral parameters. Modal analysis revealed substantial changes in fundamental period and mode shape characteristics, reflecting the enhanced stiffness contribution from the external steel framework system.

The most pronounced improvement was observed in inter-story drift control, where the exoskeleton system demonstrated exceptional effectiveness. The displacement multiplier increased from 0.758 to 3.902, representing more than a five-fold improvement in drift capacity. This enhancement corresponded to an increase in return period from 27 years to 1051 years and peak ground acceleration capacity from 0.073 g to 0.36 g.

Despite the significant improvements in displacement control, the steel exoskeleton system showed limited effectiveness in addressing strength-related failure mechanisms. Shear failure capacity improved moderately with the multiplier increasing from 0.074 to 0.546, representing substantial but insufficient enhancement. The return period for shear failure increased from 3 years to 169 years, while the corresponding PGA capacity improved from 0.027 g to 0.158 g.

More critically, flexural failure and joint failure mechanisms showed minimal improvement with exoskeleton addition. Flexural failure multiplier increased only from 0.084 to 0.234, with a return period extending from 4 years to 25 years. Joint failure response remained essentially unchanged, with the multiplier improving marginally from 0.037 to 0.067 and return period increasing from 1 year to only 1 year, indicating that local capacity limitations were not addressed by the global stiffening approach.

Detailed analysis of joint performance revealed that the most critical joints were located at the first and second floor levels, particularly at the two extremities of the structure in correspondence with the shorter beam spans. These joints exhibited the most severe capacity deficiencies due to the combination of inadequate transverse reinforcement typical of 1960s construction and unfavorable geometric conditions created by shorter spans that concentrate higher shear forces within smaller joint volumes. The limited improvement in joint performance with exoskeleton-only configuration reflects the fundamental limitation that external stiffening cannot address these localized capacity deficiencies within joint core regions. Joint shear capacity depends primarily on concrete strength and transverse reinforcement within the joint, neither of which is enhanced by external exoskeleton systems. This finding demonstrates the necessity of localized strengthening interventions to address specific capacity limitations that cannot be resolved through global stiffening approaches alone. The performance improvements achieved through exoskeleton addition are shown in

Table 9. Exoskeleton Reinforcement Seismic Performance Assessment Results.

Failure Mechanism	TR	PGA	IR,TR	IR,PGA
Exceeding chord rotation limit for R.C. columns [SLD]	9	0.044	0.419	0.376
Exceeding chord rotation limit for R.C. columns [ULS]	197	0.181	0.59	0.628
Shear failure of R.C. columns [ULS]	155	0.163	0.535	0.565
R.C. column joint failure [ULS]	0	0	0	0
Exceeding chord rotation limit for R.C. beams [SLD]	0	0	0	0
Exceeding chord rotation limit for R.C. beams [ULS]	75	0.117	0.397	0.407
Exceeding shear limit for R.C. beams [ULS]	0	0	0	0
Exceeding inter-story displacement limit [SLO] C7.8.1.5.4	7	0.039	0.466	0.424
Exceeding inter-story displacement limit [SLD] C7.8.1.5.4	9	0.044	0.419	0.376

, demonstrating the selective effectiveness across different failure modes.

6.2. Integrated System Performance

The combination of steel exoskeletons with localized CAM strengthening achieved comprehensive performance enhancement across all failure mechanisms. The integrated approach successfully addressed both global response characteristics through exoskeleton stiffening and local capacity limitations through targeted element strengthening, resulting in performance levels exceeding unity for all main parameters.

Shear failure capacity reached a multiplier of 1.009, corresponding to a 727-year return period and 0.291 g PGA capacity. This represented a 13.6-fold improvement over the baseline condition and demonstrated adequate capacity for design-level seismic forces. Flexural failure performance achieved a multiplier of 1.004 with a 719-year return period and 0.29 g PGA, indicating successful resolution of flexural capacity deficiencies through combined global and local interventions.

The biggest enhancement was observed in joint failure mechanisms, where the integrated approach achieved a multiplier of 1.025 corresponding to a 756-year return period and 0.296 g PGA. This represented a 27-fold improvement over baseline conditions, demonstrating the great importance of localized CAM strengthening in addressing joint vulnerabilities that were not effectively improved by exoskeleton addition alone.

Inter-story drift performance was further enhanced beyond the exoskeleton-only configuration, achieving a multiplier of 4.855 with a 1367-year return period and 0.395 g PGA. This improvement reflected the beneficial interaction between global stiffening and local capacity enhancement, where improved element performance enabled more effective utilization of the enhanced global stiffness characteristics. In

Table 10. Comprehensive Performance Assessment Results.

Failure Mechanism	TR	PGA	IR,TR	IR,PGA
Exceeding chord rotation limit for R.C. columns [SLD]	1643	0.36	3.545	3.06
Exceeding chord rotation limit for R.C. columns [ULS]	2475	0.395	1.667	1.367
Shear failure of R.C. columns [ULS]	75	0.117	0.397	0.407
R.C. column joint failure [ULS]	75	0.117	0.397	0.407
Exceeding chord rotation limit for R.C. beams [SLD]	2475	0.395	4.194	3.353
Exceeding chord rotation limit for R.C. beams [ULS]	2475	0.395	1.667	1.367
Exceeding shear limit for R.C. beams [ULS]	46	0.094	0.325	0.325
Exceeding inter-story displacement limit [SLO] C7.8.1.5.4	712	0.289	3.102	3.104
Exceeding inter-story displacement limit [SLD] C7.8.1.5.4	2475	0.395	4.194	3.353

, a summary of the performance assessment results for the integrated retrofit system is shown, confirming achievement of adequate safety levels across all failure mechanisms. In Figure 10 the safety check results obtained from Sismicad 13 are compared for the three case analyzed.

The integrated retrofit approach achieved a complete transformation of seismic risk classification from the baseline Risk Class G to Risk Class A+, representing the highest achievable safety level according to Italian standards. This classification change was confirmed for both PAM (Perdita Annuale Media) and IS-V (safety index) criteria, indicating comprehensive risk reduction across all evaluation parameters.

The PAM classification improved from Class G (>8.5% expected annual loss) to Class A+ (<0.5% expected annual loss), representing a reduction in expected seismic losses by more than an order of magnitude. The IS-V classification similarly improved from Class G (safety index < 0.1) to Class A+ (safety index > 4.5), demonstrating achievement of safety levels well above minimum code requirements.

This risk classification improvement confirmed the effectiveness of the integrated retrofit strategy in achieving comprehensive safety enhancement suitable for the strategic importance of educational facilities within the community infrastructure system.

6.3. Component Contribution Analysis

The systematic comparison of retrofit components revealed distinct and complementary performance contributions. Steel exoskeletons demonstrated exceptional effectiveness for displacement control, providing the primary mechanism for inter-story drift limitation while contributing moderately to overall shear capacity enhancement. However, exoskeletons showed limited capability for addressing local capacity deficiencies, particularly in joint regions and flexural capacity limitations.

Localized CAM strengthening proved essential for achieving adequate performance across strength-related failure mechanisms. The comparison between exoskeleton-only and integrated system performance clearly demonstrated that acceptable safety levels (multipliers >1.0) could not be achieved without targeted local strengthening interventions, despite the substantial global improvements provided by exoskeleton systems.

The integrated system demonstrated performance levels that exceeded simple additive effects of individual interventions, indicating beneficial synergistic interactions between global and local strengthening strategies. The enhanced local capacity provided by CAM strengthening enabled more effective utilization of the increased global stiffness from exoskeletons, resulting in optimized overall structural response.

Quantitative evidence of synergistic behavior is demonstrated by comparing expected versus observed performance improvements. If interventions were purely additive, the combined system multiplier for inter-story drift would approximate the sum of individual improvements. However, the integrated system achieves a multiplier of 4.855, which exceeds the exoskeleton-only contribution (3.902) by margins (24% improvement) that significantly exceed the modest local strengthening effects on displacement capacity alone. This enhanced performance reflects improved utilization of global stiffness through increased local element capacity, enabling more effective force distribution and system optimization.

The integrated retrofit design successfully achieved all performance objectives established for the project, with safety indices exceeding unity for all relevant failure mechanisms and return periods extending well beyond minimum code requirements. The achievement of Risk Class A+ classification confirmed compliance with enhanced performance standards appropriate for strategic educational facilities.

Pushover analysis results demonstrated robust performance margins with ultimate capacity levels providing substantial reserves beyond design-level demands. The enhanced ductility characteristics achieved through CAM strengthening contributed to improved energy dissipation capacity and reduced vulnerability to unexpected loading conditions or material degradation over the building’s service life.

The comprehensive performance validation confirmed the effectiveness of the integrated retrofit approach for addressing the complex vulnerability patterns typical of pre-seismic code construction while achieving safety levels suitable for continued building operation without occupancy restrictions or enhanced maintenance requirements.

6.4. Economic Analysis

The economic assessment of the integrated retrofit strategy employed detailed cost analysis based on the 2025 official price list for construction works in the Abruzzo region (Prezziario 2025 delle opere edili della regione Abruzzo). This approach ensured accurate cost estimation reflecting current market conditions and regional construction practices. The analysis was structured to provide transparency in cost allocation across different intervention components, enabling evaluation of cost-effectiveness for individual retrofit strategies and their combined implementation.

Cost calculations were developed through systematic quantity take-offs for all retrofit components, including material quantities, fabrication requirements, and installation procedures. Labor costs were incorporated through standard price list items reflecting typical construction productivity rates and regional wage scales. The comprehensive cost analysis included both direct construction costs and associated professional services, providing a complete project cost assessment suitable for budget planning and feasibility evaluation.

The steel exoskeleton system cost analysis was based on detailed material quantity calculations for each structural component. A single exoskeleton unit required approximately 0.96 cubic meters of steel, equivalent to 7537 kg based on standard steel density of 7850 kg/m³. The steel quantity breakdown reflected the varying profile requirements across different structural functions, with columns requiring the largest sections due to their primary load-bearing role.

Material costs were calculated using price list item ABR25_E.04.10.10.b at €3.71 per kilogram, yielding a unit cost of €27,963 per exoskeleton structure. This unit cost includes steel material, fabrication, surface treatment, transportation, and installation, providing a comprehensive cost basis for project planning. The standardized approach to exoskeleton design enabled efficient fabrication and installation procedures, contributing to cost optimization despite the custom nature of retrofit applications.

Exoskeleton installation costs included foundation preparation, structural erection, and connection to existing structure. The independent foundation system for each exoskeleton unit required excavation, concrete placement, and anchor installation, with costs integrated into the overall exoskeleton pricing through standard construction assemblies. Connection costs incorporated drilling, anchor installation, and steel connection fabrication at each interface point with the existing reinforced concrete frame.

The external installation approach minimized disruption to building operations, avoiding the premium costs associated with occupied building construction while enabling efficient use of standard construction equipment and procedures. Installation sequencing was optimized to minimize construction duration and associated indirect costs, contributing to overall project cost-effectiveness.

Column strengthening through CAM application required both concrete section enlargement and CAM system installation. The cost analysis incorporated material costs for additional concrete (0.891 m³ per column), formwork installation (19.8 m² per column), and CAM metallic strip installation. Concrete costs were calculated using price list item ABR25_E.03.10.30.a at €182 per cubic meter, while formwork costs employed item ABR25_E.03.30.10.b at €38.68 per square meter.

CAM system costs incorporated high-performance metallic strips at 0.1584 m³ per column (1243 kg), calculated at €3.71 per kilogram, consistent with structural steel pricing. The total cost per column including concrete, formwork, and CAM installation reached €5541, reflecting the comprehensive nature of the strengthening intervention. For the 32 columns requiring strengthening throughout the building, the total column strengthening cost amounted to €177,318.

Beam strengthening costs were calculated based on the total length of beams requiring CAM application. Of the total 2748 linear meters of beams within the structural system, 1557 m required strengthening based on the structural assessment results. The CAM application involved metallic strips around the beam perimeter (4 sides), resulting in 6228 linear meters of CAM installation at 0.2 × 0.02 m cross-section.

The total steel weight for beam strengthening reached 48,890 kg, calculated at €3.71 per kilogram for a total cost of €181,381. This cost reflects the extensive beam strengthening requirements identified in the vulnerability assessment, where 78 individual beam elements failed to satisfy strength requirements under seismic loading conditions.

The direct construction costs for the integrated retrofit approach totaled €1,001,841, incorporating steel exoskeleton installation (€643,141), column CAM strengthening (€177,318), and beam CAM strengthening (€181,381). This cost allocation demonstrates that steel exoskeletons represented approximately 64% of direct construction costs, while localized CAM strengthening comprised the remaining 36% despite addressing the majority of structural deficiencies identified in the baseline assessment.

The cost distribution reflects the material-intensive nature of external steel systems compared to localized strengthening applications. However, the essential role of CAM strengthening in achieving acceptable performance levels across all failure mechanisms demonstrates that both intervention types are necessary for comprehensive retrofit effectiveness.

Professional services costs were calculated using the official compensation calculator for architects and engineers according to current Italian professional standards (D.M. 17 giugno 2016 modified by D.Lgs.36/2023 [53]). The comprehensive professional services scope included preliminary design, detailed design, construction supervision, and structural testing, reflecting the complexity of integrated retrofit projects and the coordination requirements between multiple intervention strategies.

Professional services costs totaled €538,184, representing approximately 35% of direct construction costs. This proportion reflects the engineering-intensive nature of retrofit projects requiring detailed existing structure assessment, complex intervention design, and comprehensive construction oversight to ensure successful implementation and performance achievement.

The total project cost, including all construction, professional services, taxes, and contingencies, reached €2,589,550. For the 513 square meter building area, this corresponds to approximately €5050 per square meter, providing a basis for comparison with other retrofit strategies and new construction alternatives.

Cost-Effectiveness Analysis

The economic analysis demonstrates that achieving comprehensive seismic safety enhancement for existing educational buildings requires substantial investment, with costs approaching those of new construction. However, the integrated retrofit approach provides several value propositions beyond direct cost comparison, including preservation of existing architectural character, minimal disruption to educational operations, and immediate occupancy upon completion without relocation requirements.

The improvement in seismic risk classification from Class G to Class A+ represents substantial value in terms of life safety protection and asset preservation. The enhanced performance levels provide resilience against a broad range of seismic scenarios, reducing potential future losses from both structural damage and operational disruption during and following seismic events.

Comparative cost-effectiveness analysis reveals important insights regarding intervention strategies. Exoskeleton-only approaches, while representing 64% of total intervention costs, achieve inadequate performance levels (multipliers <1.0) for strength-related failure mechanisms, requiring additional future interventions. On the other hand, local strengthening alone would require treatment of 78 beam elements and 32 column elements throughout the structure, potentially exceeding the cost of the integrated approach while providing limited displacement control benefits. The integrated strategy optimizes intervention efficiency by allocating 64% of costs to global stiffening (addressing displacement concerns) and 36% to targeted local strengthening (resolving capacity deficiencies), achieving a good performance level that single approaches cannot attain.

The integrated retrofit strategy offers significant implementation advantages compared to traditional invasive retrofit approaches. The external exoskeleton installation enables work to proceed without building evacuation, maintaining educational continuity and avoiding the substantial costs and disruptions associated with temporary facility provision. The modular nature of CAM strengthening applications allows for phased implementation that can accommodate academic scheduling requirements.

These implementation advantages translate to reduced indirect costs and accelerated project delivery, enhancing the overall value proposition of the integrated approach. The non-invasive characteristics particularly benefit strategic facilities like schools, where operational continuity is essential for community function and economic activity.

The comprehensive nature of the integrated retrofit provides long-term value through enhanced building performance, reduced maintenance requirements, and improved resilience to future seismic events. The high-quality materials and construction standards employed in both exoskeleton and CAM systems ensure durability and sustained performance over extended service periods.

The achievement of Risk Class A+ performance levels provides substantial margins above minimum code requirements, offering protection against potential future increases in seismic design requirements and ensuring continued compliance throughout the building’s service life. This forward-looking approach enhances long-term value while providing confidence in sustained safety performance for this necessary community infrastructure.

The high-performance materials and construction standards employed in both exoskeleton and CAM systems could provide long-term economic advantages through reduced maintenance requirements compared to conventional retrofit approaches. Steel exoskeleton systems typically require minimal maintenance beyond periodic inspection and protective coating renewal every 15–20 years. CAM strengthening installations are designed for permanent application with no anticipated maintenance requirements throughout the building’s service life. These characteristics could contribute to favorable life cycle cost performance, though detailed LCC analysis would require long-term performance monitoring data that could be developed in future research phases.

7. Conclusions

This study quantified the performance contributions of integrated seismic retrofit strategies applied to a representative RC school building from the 1960s in Central Italy. The comprehensive assessment demonstrated that steel exoskeletons and localized CAM strengthening provide complementary but distinct structural improvements that are both necessary for achieving adequate seismic safety levels.

Steel exoskeletons proved highly effective for displacement control, improving inter-story drift capacity by more than five-fold (multiplier from 0.758 to 3.902). However, exoskeletons showed limited impact on strength-related failure mechanisms, with joint failure capacity remaining inadequate (multiplier 0.067) despite the substantial global stiffening effects. This finding challenges approaches that rely solely on external stiffening systems for comprehensive seismic retrofit of existing RC buildings.

Localized CAM strengthening proved essential for achieving performance multipliers exceeding unity across all critical failure mechanisms. The integrated approach successfully transformed the building from Risk Class G to Risk Class A+, representing complete resolution of seismic vulnerabilities through combined global and local interventions. Joint failure mechanisms showed the biggest improvement, with a relevant increase in capacity (multiplier from 0.037 to 1.025), demonstrating the importance of addressing local capacity deficiencies.

The experimental material characterization revealed significant variations in concrete properties between structural elements (C24/28 for columns vs. C20/24 for beams), confirming the importance of comprehensive testing programs for accurate assessment of existing buildings. The LC2 knowledge level achieved through systematic investigation provided appropriate confidence factors for retrofit design while demonstrating the value of investment in material characterization.

The economic analysis revealed total project costs of €2,589,550 (€5050/m²), with steel exoskeletons representing 64% of direct construction costs despite providing limited contribution to strength-related failure modes. CAM strengthening, comprising 36% of direct costs, proved essential for achieving acceptable safety levels across all performance criteria. This cost distribution highlights the importance of integrated strategies that balance global and local intervention effectiveness.

The results presented are specific to the analyzed building typology, construction period, and seismic context, as is typical for detailed structural assessments. However, the analyzed structure represents a common construction type throughout seismically active regions of Europe, and the comprehensive documentation provides valuable technical data and methodological insights applicable to the broader challenge of school building retrofit. The limitations of single case study analysis should be acknowledged, and the findings should be considered within the context of the specific structural and seismic characteristics investigated.

The analysis employed established commercial software with validated material models following current design standards. While advanced modeling approaches such as soil–structure interaction or detailed connection behavior could provide additional insights, the adopted methodology is consistent with standard engineering practice and provides reliable results for structural assessment objectives. For practitioners designing similar retrofit projects, the study demonstrates that comprehensive seismic safety enhancement requires integrated approaches addressing both global and local structural deficiencies. External stiffening systems alone, while effective for displacement control, cannot resolve the complex vulnerability patterns typical of pre-seismic code construction.

The non-invasive nature of the integrated approach enables retrofit implementation without disrupting building operations, providing significant value for strategic facilities such as schools, where operational continuity is essential. The achievement of Risk Class A+ performance provides substantial safety margins and long-term value despite the significant initial investment required.

Investment in comprehensive material characterization proves essential for accurate retrofit design and cost-effective intervention strategies. The systematic differences in material properties identified through testing directly influenced retrofit design decisions and performance predictions.

Future research could focus on several key areas to enhance the generalizability of integrated retrofit approaches: (1) parametric studies across different building geometries, construction periods, and seismic contexts to develop generalized design guidelines; (2) establishment of standardized procedures for optimal allocation between global and local strengthening strategies based on building vulnerability characteristics; (3) investigation of alternative connection strategies and standardized exoskeleton configurations to improve practical implementation; (4) experimental validation programs on representative building specimens to verify analytical predictions; and (5) development of simplified assessment tools for rapid evaluation of integrated retrofit effectiveness across the extensive inventory of vulnerable existing buildings. Multi-case comparative studies incorporating buildings with varying structural parameters, construction quality, and seismic demands would be particularly valuable for developing broadly applicable design methodologies.

Investigation of alternative connection strategies and standardized exoskeleton configurations using commercially available steel sections would improve practical implementation while potentially reducing costs. Long-term performance monitoring of completed retrofit projects would provide valuable validation data for design assumptions and performance predictions.

The development of simplified assessment tools for rapid evaluation of integrated retrofit effectiveness would support broader application of these techniques across the extensive inventory of vulnerable existing buildings requiring seismic safety enhancement.