불확실한 환경하의 FMEA에서 Fuzzy DST/AHP, TOPSIS, Grey이론을 이용한 리스크 평가 적용방안

Abstract

Failure Mode and Effect Analysis (FMEA) is a widely used method to assess the failure risk using the Risk Priority Number (RPN) of occurrence, detectability, and severity by following a tiered structure, such as the cause-mode-effect causal relationship. However, FMEA has an uncertainty under four circumstances in terms of; (1) lack of information on failure modes, (2) difference in risk criteria importance, (3) vagueness in linguistic terminologies for risk scale, and (4) limited sample number of failure mode and risk criteria. This study focuses on evaluating the failure risk in FMEA under uncertain conditions, and presents an implementation strategy of the evidential Technique for Order Preference by Similarity Ideal Solution (TOPSIS) and evidential Grey theory with group fuzzy Dempster Shafer Theory (DST)/ Analytic Hierarchy Process (AHP). Particularly, this work demonstrates the application of Basic Belief Assignment (BBA) using the weighted average with risk criteria importance and the risk scale obtained defuzzified value from corresponding linguistic terminologies. Integrated Electronic Control Unit (ECU) of Vehicle Item illustrates an implementation strategy of evidential TOPSIS and evidential Grey theory, and compares them with traditional RPN methods.