A New Failure Mode and Effects Analysis Method Based on Dempster-Shafer Theory by Integrating Evidential Network

Abstract

Failure mode and effects analysis (FMEA) is one of the most effective pre-accident prevention methods. Risk priority number (RPN) approach is a traditional method in the FMEA for risk evaluation. However, there are some shortcomings in the traditional RPN method. In this paper, we propose an FMEA approach based on Dempster-Shafer theory (DST) in an uncertainty evaluation environment. An evidential network (EN) method is proposed to establish a new model for risk evaluation in the FMEA, and we propose a novel approach to determine the conditional belief mass table (CBMT) of the non-root node. In addition, subjective weight and objective weight are integrated to determine the weights of risk factors, which can fully reflect the importance of risk factors. A numerical case is provided to illustrate the practical application of the proposed method, and the results show that this method is reasonable and effective.