A likelihood-based ORESTE method for failure mode and effect analysis (FMEA) based risk analysis problem under interval type-2 fuzzy environment

The failure mode and effect analysis (FMEA) model is an effective safety system analysis method to identify and analyze potential risks. Nevertheless, the traditional FMEA possesses deficiencies associating with questionable multiplication procedure, overlooking uncertainty and interaction relationships among risk factors. Besides, determining the risk priority considering the detailed interrelation among the failure modes is beyond the capability of current FMEA models, especially in the case of complex preference information. Thus, this paper proposed an extended FMEA by integrating ORESTE (organization, rangement et Synthese de donnees relarionnelles, in French) method, Heronian Mean (HM) operator, and interval type-2 fuzzy sets (IT2FSs). First, the IT2FSs are utilized to depict the complex risk preference information. Second, the HM operator, as an aggregated tool, is utilized to obtain collective evaluation information, which can fully reflect the overall correlations among individual risk assessment information. Then, an extended ORESTE method based on the likelihood of IT2FSs is utilized to obtain the priority ranking orders of failure modes. In this extended method, a new IT2FSs global preference score is constructed to determine the weak ranking. And, three new preference intensities based on the likelihood of IT2FSs are constructed to determine PIR (preference, indifference, and incomparability) structure which can show the detailed relationships between failure modes. Finally, an illustrative example of a nuclear reheat valve system is selected to verify its application and efficiency.