An efficient method for failure probability-based moment-independent importance measure

The failure probability-based moment-independent importance measure can well analyze the effect of input uncertainties on the failure probability of a structure or system. However, compared with the variance-based importance measure, there are few accurate and efficient methods for the computation of the moment-independent importance measure at present. In this context, a highly efficient method to compute the failure probability-based moment-independent importance measure is proposed. The proposed method estimates efficiently the conditional probability density function of the model output using the fractional moments and high-dimensional model representation-based maximum entropy method, thus the conditional failure probability can be easily obtained by integration. Finally the three-point estimation method is applied to computing the variance, namely the failure probability-based moment-independent importance measure. Since the advantages of the maximum entropy method and the three-point estimation method are inherited directly, the proposed method can yield accurate results under a small number of function evaluations. Examples in the paper demonstrate the advantages of the proposed method as compared with existing methods, and indicate its good prospect for engineering application.