A Neuro-Fuzzy Risk Prediction Methodology in the Automotive Part Industry

Failure mode and effects analysis (FMEA) is a systematic and structured method employed across diverse industries to proactively identify and evaluate potential failure modes. In a traditional FMEA, for all failure modes, three criticality parameters, severity, detection, and frequency, are assessed to evaluate criticality. Nevertheless, it frequently has certain flaws. Therefore, in this work, a fuzzy risk proposed model is used to improve the use of the FMEA methodology. The new model uses a fuzzy inference technique in place of the conventional criticality calculation. The fuzzy logic technique is used where the various factors are given as members of a fuzzy set fuzzified by employing adequate membership functions to evaluate the risk and then ranking failure modes and preferring measures to control the risks of undesired events. The adaptive neuro-fuzzy inference system (ANFIS) is suggested as a dynamic, intelligently proposed model to improve and validate the results acquired by the fuzzy inference system and effectively predict the criticality evaluation of failure modes. Finally, an automotive parts industry case is presented to show the potential of the suggested model. This analysis offers a different ranking of failure modes and improves decision-making by providing a “preventive–corrective plan.” A comparison with existing approaches is presented to demonstrate the efficiency of the suggested approach.