A probabilistic evaluation of pier-scour potential in the Gaoping River Basin of Taiwan

Abstract

A probabilistic approach is used to create a preliminary inspection evaluation form (PIEF) for scour potential at bridge sites. In Taiwan, the risk of pier scour is often evaluated using a two-step procedure. First, a bridge is visually inspected based on a PIEF. An advanced scour risk analysis is conducted for a bridge with a high PIEF score. Because a PIEF can quickly evaluate scour potential, it can be used to build a maintenance sequence for a group of bridges. However, a PIEF is often created based on only engineers experience; the accuracy and reliability of PIEFs have not been examined systematically. Thus, a probabilistic-based PIEF is constructed by establishing a close correlation between the PIEF score and scour potential via Taguchi method. The scour potential is evaluated by Bayesian network (BN) that incorporates experts judgments and results of reliability analyses. For example, the conditional probabilities (CP) in the proposed BN are calculated based on an existing PIEF and results of a stability-based reliability analysis. Thus, the proposed PIEF implicitly considers the probabilistic characteristics in the scour potential, which will provide an efficient and accurate evaluation of scour potential and assist in establishing maintenance priorities of existing bridges in Taiwan.