PSI - Issue 78

Matteo Tatangelo et al. / Procedia Structural Integrity 78 (2026) 404–411

405

1. Introduction Nowadays, so-called traditional methods of measurement of safety, in which safety factors have a deterministic character, are still widely adopted. The variables describing the construction, strength, and applied loads are taken as known values in which it is assumed there is no uncertainty. Decisions about constructions design and assessment are nowadays recognized as needing to apply risk-based criteria. This means taking into account not only loss of life and injuries, but also monetary losses and environmental damages. This approach necessitates calculating total risks throughout the entire life-cycle of the construction, relying on probabilistic models to evaluate exposures, damage, failure events, and their subsequent direct and indirect consequences (ISO 2394, 1998). The implementation of a reliability-based approach allows the optimization of costs and consumed resources, subject to the fulfilment of given reliability requirements. The genesis of these requirements should be based on a risk-based approach, depending on the consequences of failure and the costs associated with reliability improvements. In addition, it is essential to recognize that reliability requirements are, by their nature, influenced by socio-economic considerations, they may differ among the various codes (ISO 2394, 1998). In this context, this study presents a reliability-based procedure for life-cycle management of new and existing constructions (Tatangelo et al. 2024). It consists in the conversion of a time-integrated into a time-dependent approach evaluating, within a certain time interval, the estimated reliability to be compared with the target one for a given limit state. In this way, design of new constructions and assessment of the existing ones fall within an unitary methodology regarding the entire life-cycle. Therefore, more structured and gradual construction management is now possible, identifying different thresholds of attention as the difference between capacity and demand decreases. The procedure is applied to a case study, consisting of RC half-joints, also known as Gerber saddles. The latter are critical elements in existing RC bridges, as they introduce structural discontinuities that may accelerate deterioration due to infiltration of aggressive agents such as chlorides and water. Although their isostatic configuration reduces internal restraint forces, it also increases the risk of progressive collapse if a single element fails. Therefore, it is interesting to analyze their structural reliability during the life-cycle, considering the evolution of loads and material degradation. 2. Target reliability indexes In a reliability-based approach, the acceptance criteria are defined as the target reliability. This is the minimum reliability needed to ensure a construction, or its components, remain safe and serviceable over a specific reference period (JCSS PMC, 2001). Reliability targets are set through an optimization process designed to lower expected losses, including economic, human, and social impacts, to acceptable levels (JCSS PMC, 2001). For example, a structural failure requires considering both direct costs, like repairs or replacement, and indirect consequences such as social disruption, occupant displacement, injuries, fatalities, and environmental effects from CO2 emissions, energy use, or the release of hazardous materials (ISO 2394, 1998). Target reliability is typically represented by an index, which corresponds to a conventional probability of failure ( , ). Currently, many standards and technical documents suggest target reliability indexes for a given , considering its Consequence Class (CC) and Limit State (LS), such as Ultimate Limit State (ULS), Fatigue Limit State (FLS), and Serviceability Limit State (SLS). Figure 1 compares various target reliability indexes, illustrating them for different CCs and LSs. It identifies between new and existing constructions and highlights the appropriate reference periods. The comparison clearly reveals a significant discrepancy among the target reliability values provided. This inconsistency derives mainly from differing definitions of CCs and varied considerations (with target values typically indicated as 1 and 50 ). Despite these differences, the target values generally fall within the range of 0 to less than 6. In fib Bulletin n. 80 (2016) offers a detailed reliability-based approach for existing bridges and RC structures. Unlike earlier documents, it allows calculating target reliability indexes for a specific constructions, even if differ from the nominal life of a new construction.