PSI - Issue 84

Simone Celati et al. / Procedia Structural Integrity 84 (2026) 127–134

133

Fig. 1: Chloride content distributions (prior left and likelihood right) at cover depth for years 10, 20, 50 and 100.

Fig. 2: Prior and predictive distributions of the time to initiation variable

4.5. Updating of the reliability of the pier caisson with the results from chloride ingress tests The structural reliability is updated based on the updated time to corrosion initiation ( ℎ ) model as described in Section 4.3. The time-dependent reliability of the considered caisson subjected to ship impact load is evaluated as described in Sections 4.1 to 4.3, using the updated distribution for the time to initiation. The updated distribution is described using a kernel distribution with the updated mean and standard deviation equal to 106 and 64 years, respectively. The updated time-dependent reliability index ( , , see also Eq. (14)) is shown in Fig. 3. The figure shows that the reliability index is only slightly affected by the results of the chloride ingress tests, despite these tests revealing a much higher concentration of chloride ions in the concrete than what was predicted by the prior model. From Fig. 3 it can be observed that the reliability index begins to decrease after approximately 50 years, i.e., when the survival probability function for the time to initiation drops to around 80%. This observation highlights that corrosion initiation does not immediately imply structural failure. Rather, the structure's reliability is only affected by the corrosion process once the probability of initiation reaches relatively high levels (around 20%). 5. Summary, conclusions and outlook In this paper, we have introduced a probabilistic approach for the quantification of structural reliability in reinforced concrete structures using chloride concentration measurements. The proposed approach explicitly incorporates uncertainties associated with both the chloride ingress prediction model and the measurement process, allowing for reliability updating based on observed data. A key feature of the approach is the preservation of temporal characteristics in the chloride transport process. This is achieved by updating both the depth-specific apparent chloride concentrations and the time-specific diffusion coefficients at the time of measurement. Through the presented case study, we demonstrated how chloride concentration measurements influence structural reliability, showing that both the measured chloride concentrations and the updated uncertainties affect the updated structural reliability. Interestingly, we observed that higher measured chloride concentrations do not necessarily lead to a significant decrease in structural reliability, stemming from the interaction of measurement data and the uncertainty reduction. Furthermore, we explored a link between corrosion initiation probability and structural reliability, providing an interpretation guidance towards the structural reliability impact. The proposed methodology represents a first step towards a more rational and data-informed treatment of corrosion-induced degradation in reinforced concrete