A hybrid multi-criteria decision-making approach for the risk assessment in the automotive parts industry

Abstract

Failure modes and effects analysis (FMEA) is a structured method frequently utilized to recognize and eliminate possible failure modes. In the overall FMEA procedure, failure modes are evaluated and prioritized based on the risk priority numbers (RPN), calculated by multiplying three parameters: frequency (F), non-detection (ND), and severity (S). However, the conventional FMEA method fails to cope with the uncertainty in complex systems. To address the challenge of ambiguity and uncertainty, an Adaptive Neuro-Fuzzy Inference System (ANFIS) is used as a dynamic selected model where these factors are represented as components of a fuzzy set that has been fuzzified utilizing the suitable membership functions to predict the criticality evaluation of failure modes. In addition, to enhance decision-makers' confidence a new hybrid model for acquiring a more logical ranking of failure modes is suggested. Essentially, two parts of the evaluation procedure are explained: determining the weights of risk parameters using the Fuzzy Analytic Hierarchy Process (FAHP) and ranking the failure modes using a mathematical rough set theory and ‘Technique for Order Performance by Similarity to Ideal Solution’ (TOPSIS) method. In the current study, the failure modes are ranked by considering two further parameters, involving cost and treatment duration. Finally, an automotive parts industry case is presented to show the potential of the proposed approach. This analysis provides a different ranking of failure modes and improves the decision-making by providing a “preventive –corrective plan. A comparison with existing approaches is presented to demonstrate the efficiency of the suggested approach.