PSI - Issue 84

Vittorio Palma et al. / Procedia Structural Integrity 84 (2026) 630–637

636

conveyance), Pozzetti intasati (clogged drains), Scarichi corti (short drains), Scarichi ostruiti (obstructed drains), Scarichi danneggiati (damaged drains) and Distacco tampone (detached buffer) (Ministero delle Infrastrutture e dei Trasporti (MIT) 2020). This results in a final classification that ranks the structures primarily according to their defectiveness level (from highest to lowest) and, secondarily, by the number of durability-related defects (from highest to lowest). In this updated version, the prioritization of works is integrated with a classification dedicated exclusively to the parameter of user safety. This implementation responds to the operational need of managers to intervene promptly on critical issues which, while not compromising the overall stability of the structure, require a targeted solution to protect public safety. The ultimate goal is to obtain a descending priority list, placing at the top those works that have the highest number of user safety reports recorded by inspectors during periodic campaigns. This approach makes it possible to isolate issues related to the potential fall of materials, even where there are no primary structural defects, ensuring that surface restoration work is planned completely independently and separately from long-term strategic investment planning. This ensures a rapid and effective response to the need to secure areas below or adjacent to the infrastructure. 6. Discussion on the methodological framework and scope of application The proposed methodologies provide an operational formalisation of the programming process prescribed by the Guidelines, supporting network-level decision-making under conditions of heterogeneous information and limited resources. The contribution of the work is primarily methodological: rather than introducing new indicators, the framework explicitly structures the decision rules that link inspection outcomes and classification results to operational priorities, formalising procedures that in practice are often applied implicitly or in a non-uniform manner. Consistency with the Guidelines is preserved throughout the entire decision-making pathway. Each module is embedded within a specific phase of the multi-level process, maintaining its normative and informational assumptions: special inspections are prioritised based on the vulnerability of the post-tensioning system, safety assessments are driven by the presence of critical conditions and by structural–foundation defectiveness, and local interventions are addressed within a distinct management domain focused on degradation mitigation and user safety. The proposed framework therefore does not introduce a new taxonomy, but clarifies the transition from technical classification to activity programming already implicit in the regulatory structure. A key feature of the approach is the systematic adoption of non-compensative criteria, whereby the guiding variable of each module governs the assignment of the main priority class. This reflects a precautionary principle typical of existing infrastructure management, preventing locally critical conditions from being mitigated by favourable contextual or exposure indicators. Secondary criteria are used exclusively for within-class refinement, ensuring interpretable rankings and reducing the risk of counter-intuitive results often associated with weighted-score approaches. The framework is explicitly designed for network-level programming phases, where available information is predominantly inspection-based, classificatory, and semi-quantitative, and does not support reliable probabilistic safety assessments. Probabilistic and reliability-based approaches remain the reference for detailed safety evaluation, but require levels of knowledge, modelling effort, and calibration that are generally feasible only for selected assets or in more advanced phases of the assessment process. Within this context, the proposed methodology is not an alternative to probabilistic models, but a complementary tool aimed at directing the acquisition of further knowledge toward those assets for which advanced analyses are most justified. The benefits of the proposed approach are particularly evident for large bridge networks characterised by heterogeneous information and the need for periodic priority updates. The use of ordinal variables and explicit decision rules enables application under incomplete knowledge, while the representation through discrete classes, matrices, and decision schemes facilitates implementation within asset management systems and supports transparency and verifiability of decisions. The explicit consideration of exposure as a refinement criterion further allows functional consequences on the transport network to be addressed without conflating them with structural safety assessment. Intrinsic limitations remain, mainly related to the quality and consistency of inspection data and to the definition of operational thresholds and combination matrices. While these elements support scalability, they may require calibration to specific network contexts and updates as the information base evolves. Nevertheless, the proposed formalisation provides a stable decision structure within which the effects of different classification assumptions and methodological refinements