PSI - Issue 84

Angelo Masi et al. / Procedia Structural Integrity 84 (2026) 321–328

322

1. Introduction Bridges represent a fundamental component of transportation networks, ensuring continuity of mobility and supporting economic and social activities. However, a large portion of the existing bridge stock has reached or exceeded its original design life, making systematic inspection and maintenance activities increasingly crucial to guarantee safety and serviceability (Ellis, 2024). Recent catastrophic failures, such as the collapse of the Polcevera viaduct, have further highlighted the consequences of inadequate management strategies for ageing infrastructures (Bazzucchi et al., 2018). In response to these issues, Italy introduced a comprehensive regulatory framework for bridge management through the Italian Guidelines for risk classification and management, safety assessment, and monitoring of existing bridges (LG2020) (MIT, 2020) (Cosenza & Losanno, 2021), in line with the Italian Building Code (MIT, 2018). Updated in 2022 (MIT, 2022), LG2020 adopts a multilevel approach to bridge risk assessment (Santarsiero et al., 2021), where Level 1 inspections require the systematic compilation of defect forms for each structural component. The overarching objective is to establish a homogeneous bridge management system among different road authorities and to support consistent, risk-informed decision-making processes (Brighenti et al., 2024). Within this framework, bridge bearing devices play a key structural role, as they transfer loads from the superstructure to the substructure while accommodating imposed displacements and rotations due to traffic, temperature variations, and environmental actions (Marioni, 1983). Moreover, these components are particularly numerous; for example, in common girder bridges their number is typically about twice that of the beams, requiring inspectors to complete a very large number of inspection forms. Bearings are widely recognised as failure-critical components, similarly to other vulnerable elements such as half joints (Santarsiero et al., 2025), because their malfunction can directly compromise the global safety and functionality of the bridge (Lee, 1994). Over time, bearing devices are exposed to deterioration mechanisms driven by ageing, environmental aggressiveness, and repeated loading, potentially leading to severe performance degradation. For these reasons, the assessment of bearing condition is essential for both structural safety and seismic performance (Ozsarac, 2023). Simplified approaches for preliminary verification of existing bearings have been proposed in the literature, for example by comparing design actions prescribed by outdated and current standards to derive performance indices (Santarsiero et al., 2023). Nevertheless, LG2020 places particular emphasis on visual inspection activities, which remain affected by subjectivity and variability among inspectors. Differences in experience, assessment criteria, and defect interpretation may lead to inconsistencies in condition evaluation, ultimately reducing the reliability of maintenance planning (Rossi et al., 2023). To mitigate these issues, recent studies have explored automated approaches based on artificial intelligence for bearing recognition and damage detection, although their effectiveness strongly depends on the availability of high-quality annotated datasets produced by experienced engineers (Liang et al., 2024). At present, LG2020 is being applied in the final stage of an experimental phase (ending in December 2025) with the support of the ReLUIS Consortium, involving a large number of universities and research institutions across Italy. Within this context, the systematic collection and analysis of inspection data represent a valuable opportunity to assess the effectiveness of the current procedures and to identify potential limitations requiring refinement. Exploiting data collected during the large-scale application of LG2020 to an extensive bridge stock, this study focuses on the development of a structured database of bridge bearings and presents a preliminary analysis of bearing typologies and observed defects, aiming to support future improvements in inspection practices and infrastructure management strategies. This paper, therefore, briefly introduces the database structure and a few observations related to inspection procedures and challenges based on a preliminary analysis. In particular, Section 2 presents the inspection procedure and the database structure, Section 3 shows the analysis results in terms of bearing typologies and main defectiveness, while Section 4 reports possible challenges in the whole bearing defect assessment according to LG2020. 2. Bearing Devices, Inspection Framework and Database Structure Bridge bearing devices are key structural components responsible for transferring vertical and horizontal actions from the superstructure to the substructure, while allowing the displacements and rotations required to accommodate thermal effects, traffic loads, time-dependent phenomena, and seismic actions (Lee, 1994). Depending on their kinematic function, bearings may provide fixed or movable horizontal restraints, whereas vertical stiffness is always