An Improved Failure Mode and Effects Analysis Method Using Belief Jensen–Shannon Divergence and Entropy Measure in the Evidence Theory

Failure mode and effects analysis (FMEA) is a typical risk assessment and prevention technology. It works by integrating different assessments provided by different experts to determine the risk level of the target FMEA item. However, in the process of using FMEA method, when experts provide subjective or inaccurate assessments, there is uncertainty. In addition, when several experts give different or even conflicting assessments, there is conflict. Thus, how to deal with the uncertainty and conflict in the FMEA process is an open problem. In this work, a new risk priority model based on the belief Jensen–Shannon divergence measure and Deng entropy is proposed. In the new method, Deng entropy is used to model the uncertainty of risk assessments, and the belief Jensen–Shannon divergence measure is used to deal with potential conflict information and neutralize abnormal assessments. Dempster’s combination rule will be used to fuse data to generate integrated values of the risk factors. Finally, the practicability and effectiveness of the new method are verified by a case study on the sheet steel production process.