Seismic resilience of a bridge based on fuzzy-probabilistic approach

In this paper, a new model based on a fuzzy-probabilistic approach for predicting seismic resilience of a bridge is developed. The main purpose is to use seismic resilience in decision making process for disaster management during the pre-disaster period. Another aim is to include contingency in resilience assessment. In previous research, uncertainties have not been fully considered in the proposed models for seismic resilience assessment. Since the residual functionality of bridges depends on its vague damage states, so it is presented by fuzzy triangular numbers. However, the both idle time interval and recovery duration are random in nature, so they are described as random variables. The Monte Carlo simulation is used for generating 10000 samples of these variables. Further, resilience is represented as fuzzy-random variable with corresponding fuzzy mean value and fuzzy standard deviation obtained from the generated and estimated data. The functionality of the system is therefore described as a fuzzy-random function whose shape depends on the disaster preparedness of the system. The resilient curves are illustrated using fuzzy functions. A Java application is developed for purpose of resilience assessment. For a case study, a bridge in Santa Barbara is chosen. The result of resilience assessment process is used for decision making in disaster management and emergency responses.