Enhancing reliability of failure modes and effects analysis dealing with linguistic distribution assessments: A consistency based approach
Xiang Jia,Ying-Ming Wang,Luis Martínez
DOI: https://doi.org/10.1016/j.engappai.2024.108333
IF: 8
2024-04-01
Engineering Applications of Artificial Intelligence
Abstract:Failure modes and effects analysis is a proactive reliability-management engineering technique, it has been applied to improve reliability of processes, products, and services. Traditional failure modes and effects analysis method manifests three deficiencies with regard to form of risk assessments, determination of risk factor weights, and exploitation of failure modes' risk levels. Although numerous studies have contributed to surmount these deficiencies, existing failure modes and effects analysis methods still face such issues. Moreover, a novel drawback has raised when failure modes and effects analysis participants adopt linguistic terms as risk assessments of failure modes, the reliability of linguistic distribution assessments as well as failure modes and effects analysis groups (participants with similar knowledge background are previously classified into a failure modes and effects analysis group) is overlooked. Therefore, this paper introduces the transformation of linguistic distribution assessments into fuzzy preference relations, as more effective modeling for investigating consistency, by using the newly defined possibility degrees of linguistic distribution assessments will be applied to devise four algorithms for enhancing reliability of assessments. Hence, the three deficiencies of traditional failure modes and effects analysis method are solved by means of a novel failure modes and effects analysis framework. Finally, a real example with respect to ranking failure modes of worm wheel grinding machine is developed to verify the performance and advantages of the proposed method, in which ten failure modes are ranked and the one with the highest risk level is selected, and consistency of risk assessments is improved to be acceptable.
automation & control systems,computer science, artificial intelligence,engineering, electrical & electronic, multidisciplinary