Importance Analysis Based on Evidence Theory and Bayesian Network Reasoning

To address the deviation in importance analysis caused by uncertainty in evaluating the occurrence probability of failures with limited sample size, we propose an importance analysis method based on evidence theory and Bayesian network reasoning. The method utilizes Structured Analysis and Design Technique (SADT) for system function analysis and integrates Failure Mode and Effect Analysis (FMEA) to construct a Bayesian network (BN) structure model. We employ evidence theory to integrate multi-source information and calculate the failure rate interval of system components. Then realized the importance analysis of system components by Bayesian reverse reasoning, which provides guidance for improving the reliability of the system. To illustrate the power of our method, we present a case study involving a certain series of machining center chain type tool magazine and Automatic Tool Changer (ATC).