Risk Assessment of Bauxite Maritime Logistics Based on Improved FMECA and Fuzzy Bayesian Network

Abstract

Because of the many limitations of the traditional failure mode effect and criticality analysis (FMECA), an integrated risk assessment model with improved FMECA, fuzzy Bayesian networks (FBN), and improved evidence reasoning (ER) is proposed. A new risk characterization parameter system is constructed in the model. A fuzzy rule base system based on the confidence structure is constructed by combining fuzzy set theory with expert knowledge, and BN reasoning technology is used to realize the importance ranking of the hazard degree of maritime logistics risk events. The improved ER based on weight distribution and matrix analysis can effectively integrate the results of risk event assessment and realize the hazard evaluation of the maritime logistics system from the overall perspective. The effectiveness and feasibility of the model are verified by carrying out a risk assessment on the maritime logistics of importing bauxite to China. The research results show that the priority of risk events in the maritime logistics of bauxite are “pirates or terrorist attacks” and “workers’ riots or strikes” in sequence. In addition, the bauxite maritime logistics system is at a medium- to high-risk level as a whole. The proposed model is expected to provide a systematic risk assessment model and framework for the engineering field.