This paper presents a time-dependent reliability assessment method for corroded steel-concrete composite (SCC) beams using the combined Bayesian updating algorithm and Monte Carlo simulations (MCS). A corrosion model is first developed to describe the degradation of the beam’s cross-sectional area over time. The corrosion depth is characterized as a stochastic process based on previously proposed models and empirical data. Bayesian updating is then applied to continuously refine the probabilistic description of uncertain parameters and the system’s performance as new information becomes available. This enables a more accurate evaluation of failure probability throughout the service life of the structure. Three environmental conditions, which are rural, urban and marine areas, are considered in reliability evaluation of corroded SCC beams. A sensitivity analysis is conducted to quantify the influence of model parameters on structural reliability, thereby supporting informed decisions in structural design and maintenance planning. Finally, the proposed approach is benchmarked against traditional reliability methods, including MCS, First-Order Reliability Method (FORM), and Second-Order Reliability Method (SORM). The results demonstrate that the Bayesian updating framework provides improved accuracy, especially under limited data conditions, and facilitates the efficient incorporation of new information into the time-variant reliability analysis.
Nguyen et al. (Thu,) studied this question.