A novel failure mode and effect analysis method using a flexible knowledge acquisition framework based on picture fuzzy sets. (January 2023)

• Record Type: Journal Article
• Authors: Zhao, Xiang-Kun; Zhu, Xiao-Min; Bai, Kai-Yuan; Zhang, Run-Tong
• Is Part Of: Engineering applications of artificial intelligence. Volume 117:Part A(2023)
• Publication Date: 2023-01
• Subjects: Failure mode and effect analysis -- Evidential reasoning -- Knowledge acquisition -- Picture fuzzy sets; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2022.105625
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Failure mode and effect analysis (FMEA) is an effective reliability management tool for identifying potential failures in a system/component that has been widely utilized in various fields by combining fuzzy sets theory. However, the difficulty in assessment expression and acquisition, imprecision in assessment aggregation, and missing relationships among risk factors are prominent challenges in the existing fuzzy FMEA methods. Thus, this work employs the picture fuzzy sets (PFSs) theory to meet these challenges, allowing experts to express assessments more efficiently and accurately. Meanwhile, we propose a novel FMEA method to improve the three essential processes of FMEA, involving the following steps. Firstly, a flexible knowledge acquisition framework (FKAF) is established to simplify the expert evaluation process, allowing experts to express fuzzy information with various forms of non-fuzzy assessments. Then, a strategy-based picture fuzzy conversion (PFC) method is developed to standardize the non-fuzzy values to picture fuzzy numbers (PFNs). Secondly, to improve the assessment aggregation accuracy in uncertain environments, we suggest the picture fuzzy evidential reasoning (PFER) method that extend the existing fuzzy evidential reasoning (FER) methods. Thirdly, to completely describe the parallel and causal relationships among risk factors, four alternative models are established using picture fuzzy Petri nets (PFPNs) and risk priority rankings are determined through inference. Finally, the effectiveness and superiority of the proposed FMEA method are verified through two case studies and one extended experiment, demonstrating that it overcomes the shortcomings of the existing FMEA methods and reduces application costs while ensuring the rationality of rankings.