Failure Mode and Effects Analysis Using Variable Precision Rough Set Theory and TODIM Method

摘要

Failure mode and effects analysis (FMEA) is a widely used tool of risk assessment to identify and eliminate failures of products and systems. However, the conventional FMEA has some defects, such as the same importance of risk factors, utilizing crisp numbers to evaluate failures without considering vagueness. Although fuzzy methods are used to improve the conventional FMEA, the fuzzy FMEA requires much priori information (e.g., fuzzy membership function), which cannot flexibly reflect the changes of a decision makers preference. In addition, most of the previous methods suppose that decision makers are totally rational without considering their psychological factors. Actually, FMEA members' judgments on the failure modes are often influenced by their bounded rationality. To solve those problems, an integrated FMEA model is proposed in this paper, which integrates the strength of variable precision rough set theory in handling vagueness and the merit of TODIM (an acronym in Portuguese of Interactive and Multicriteria Decision Making) approach in manipulating bounded rationality of decision makers. Finally, the proposed method is validated with a case study of a steam valve system to demonstrate its effectiveness and efficiency.