Exploring the Ordinal Classifications of Failure Modes in the Reliability Management: An Optimization-Based Consensus Model with Bounded Confidences

Failure mode and effect analysis (FMEA) is a system activity that identifies, evaluates and eliminates potential failure modes (FMs) in a system/process to enhance the quality and reliability of a product. In order to improve the implementation efficiency of FMEA, this study proposes a consensus-based FMEA method to derive the ordinal classifications of FMs, in which the FMEA team members employ linguistic distribution to convey their preferences. In the proposed FMEA method, a multi-stage consensus optimization model with bounded confidences is designed to help the FMEA team reach a consensus. In the consensus reaching process, a maximum consensus optimization model based on bounded confidences is provided to obtain the adjustment suggestions by maximizing the level of consensus among the FMEA team. If the predetermined level of consensus cannot be reached, the adjustment suggestions obtained by the maximum consensus optimization model are adopted to guide the preference-modification of the FMEA team members. Otherwise, a two-stage consensus optimization model based on bounded confidences is designed to derive the adjustment suggestions for the preference-modification of the FMEA team members. Finally, a case study of marine diesel engine crankcase explosion, a sensitivity analysis and a comparative analysis are proposed to illustrate the feasibility and effectiveness of the proposed FMEA method.

Automated summary

This study proposes a consensus-based FMEA method to derive ordinal classifications of failure modes using linguistic distribution, and designs a multi-stage consensus optimization model to help the FMEA team reach consensus.