Reliability Analysis Method Study of Multi-state System Based on Fuzzy Bayesian Network

ABSTRACT

To deal with fuzzy and uncertain information in reliability analysis of multi-state systems, an approach based on fuzzy Bayesian networks is proposed. In this approach the fuzzy set theory and Bayesian networks methodology are applied. The fuzzy assessment information from different experts is described by triangular fuzzy numbers, which denotes the fuzzy probabilities of different states of root nodes. The depended uncertain ordered weighted averaging (DUOWA) operator is used to aggregate assessments information. Through defuzzification calculation, the crisp probabilities are obtained as the prior probabilities of the root nodes. Bayesian network (BN) is used to handle the uncertainty of logical relationship between nodes and to compute system reliability. A case study concerning the reliability analysis in cell production line is provided and conducted using the proposed approach to illustrate its effectivity and practicability.

Keywords: