Reliability-Based Analysis of a Portuguese Concrete Arch Bridge Employing Surrogate Modeling Techniques Based on Adaptive Sequential Sampling

Adaptive sequential sampling provides a good technique to refine and increase the accuracy of surrogate models used for reliability analysis based on selecting possible future candidates in the input domain (i.e., random variables). In the present research, adaptive sequential sampling was employed to obtain the training experimental design from a set of random variables to develop a surrogate model capable of representing the failure limit of the asset under vertical traffic loads. The surrogate model was then used to obtain the failure probability of the case study, a single-span concrete arch bridge in Portugal. The resulting reliability index was compared with the outcomes obtained in previous research. Finally, by integrating novel sampling approaches a reduction in the computational cost associated with the reliability analysis and an overall better reliability of the surrogate model for evaluating the performance limit state was obtained.