Risk prioritization in Failure Mode and Effects Analysis using interval type-2 fuzzy sets


Bozdag E., Asan U. , SOYER A. , SERDARASAN S.

EXPERT SYSTEMS WITH APPLICATIONS, vol.42, no.8, pp.4000-4015, 2015 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 42 Issue: 8
  • Publication Date: 2015
  • Doi Number: 10.1016/j.eswa.2015.01.015
  • Title of Journal : EXPERT SYSTEMS WITH APPLICATIONS
  • Page Numbers: pp.4000-4015
  • Keywords: Failure Mode and Effects Analysis, Interval type-2 fuzzy sets, Uncertainty, EVIDENTIAL REASONING APPROACH, CRITERIA DECISION-ANALYSIS, TOPSIS APPROACH, OPERATORS, 2-TUPLE, OWA

Abstract

The analysis of failure modes and their effects generally requires dealing with uncertainty and subjectivity inherent in the risk assessment process. A review of the literature reveals that although a number of studies have examined these issues, none of them have explicitly studied the variation in one expert's understanding (intra-personal uncertainty) and the variations in the understanding among experts (inter-personal uncertainty) together. To address this problem, this paper proposes a new fuzzy FMEA approach based on IT2 fuzzy sets, which has the ability to capture both intra-personal and inter-personal uncertainty. The approach introduces three methods that are new for the analysis of failure modes. First, to provide a more accurate representation of the aggregated data by preserving the variations among the individual judgments a new aggregation method is suggested. It transforms individual judgments in form of intervals into a group judgment in form of an IT2 FN. Second, to allow considering optimal weights for the risk factors and thereby developing more flexible structures for their synthesis, an a-cut based ordered weighted averaging operator is adapted. Finally, to rank failure modes on a continuous scale and reflect subtle differences in the assessments properly, a ranking method for IT2 FNs based on a-cuts is adopted. The applicability and effectiveness of the proposed approach is demonstrated by an illustrative example. Comparisons with the results of crisp- and fuzzy-based methods demonstrate that the proposed approach offers additional flexibility to the experts in making judgments and provides a better modeling of uncertainty. (C) 2015 Elsevier Ltd. All rights reserved.