PSI - Issue 78

Simone Reale et al. / Procedia Structural Integrity 78 (2026) 1657–1664

1663

Where |dλ(IM>x)| is the absolute value of the derivative of the site hazard curve. For consistency with the previous section, the discussion will refer to the λ f values related to DS4. Figure 1 reports the evolution over time of λ f and the normalized values with respect to the initial λ f value. λ f values computed applying Strategy 1 are almost constant over time, due to the oversimplified representation of corrosion damage. It is worth noting that seismic risk associated with Strategy 4 is higher also for the initial time due to the inclusion of bond-slip effects. λ f values resulting from Strategies 2, 3 and 4 denote an increase of the seismic risk metric up to around 1.2, 1.36 and 1.66 times the initial value. It is therefore clear that Strategy 1 is not suitable for long-term seismic performance assessment of RC bridges. The results confirm that a more detailed description of corrosion damage is associated with more pronounced effects on the long term seismic assessment. It is worth noting that the results have been obtained by the analysis of a modern bridge and that the same exercise repeated on existing bridges may lead to even more pronounced discrepancies between the different approaches. Finally, the effects of corrosion on phenomena such as LCF and bond-slip still represent open research questions, possibly hindering the practical application of Strategy 4. The assumptions on the level of detail of corrosion damage modeling are fundamental for the assessment, highlighting the need for additional research.

Figure 1. (a) λf values and (b) normalized λf values over time for the four corrosion damage modeling strategies.

5. Conclusions This contribution investigated the implications of four progressively more refined corrosion damage modeling strategies on the long-term seismic performance assessment of RC bridges considering a case study bridge exposed to chloride attack. The assessment has been carried out according to the PBEE principles. Corrosion evolution over time has been characterized through Monte Carlo simulation accounting for the inherent uncertainty of the process. The results revealed that the median IM values of the fragility curves tend to decrease over time, in a more pronounced way as more details are included in the corrosion damage modeling strategies. Risk values computed through Strategy 1 are almost constant over time, while Strategy 2, 3 and 4 denote an increase of the seismic risk metric up to around 1.2, 1.36 and 1.66 times the initial value. Strategy 1 appears to not be suitable for long-term seismic performance assessment of RC bridges. The results confirm that a more detailed description of corrosion damage is associated with more pronounced effects on the long-term seismic assessment, but additional research is needed on this issue given that the practical application of more refined strategies may be hindered by the lack of experimental validation. Acknowledgments The first author acknowledges the financial support provided through the doctoral program at University of Pavia.