An efficient method based on Bayes' theorem to estimate the failure-probability-based sensitivity measure

The failure-probability-based sensitivity, which measures the effect of input variables on the structural failure probability, can provide useful information in reliability based design optimization. The traditional method for estimating the failure-probability-based sensitivity measure requires a nested sampling procedure and the computational cost depends on the total number of input variables. In this paper, a new efficient method based on Bayes' theorem is proposed to estimate the failure-probability-based sensitivity measure. The proposed method avoids the nested sampling procedure and only requires a single set of samples to estimate the failure-probability-based sensitivity measure. The computational cost of the proposed method does not depend on the total number of input variables. One numerical example and three engineering examples are employed to illustrate the accuracy and efficiency of the proposed method.