Probabilistic seismic safety assessment of bridges with random pier scouring

Foundation scour has been a reason for several cases of river-bridge earthquake-induced failure during recent decades. However, practising engineers often do not consider its direct effect on the seismic design procedure of such structures. The cavity around a bridge foundation is a random phenomenon depending on several uncertain parameters. This study provides a probabilistic platform to investigate the effect of random scouring on the seismic performance of a particular bridge. The procedure is implemented on an existing multi-span reinforced concrete bridge. To this end, the Monte Carlo simulation technique is utilised to generate samples of the random variables of the scour model to develop the scour hazard curve. A common type of reinforced concrete multi-span bridge is considered as a model. The Latin hypercube sampling method is employed to generate random scouring scenarios in the finite-element model, including uniform and non-uniform scour. Then, fragility curves are developed utilising cloud dynamic analysis. The results reveal that the scouring pattern is one of the most crucial sources of uncertainty. In most circumstances, uniform scour scenarios are more effective than the average of non-uniform cases. However, in some specific patterns, the effect of non-uniform scouring is dominant.