mode and effects analysis (FMEA) is an effective tool to assess the risk of a system or process under uncertain environment. However, how to handle the uncertainty in the subjective assessment is an open issue. In this paper, a novel method to deal with the uncertainty coming from subjective assessments of FMEA experts is proposed in the framework of Dempster–Shafer evidence theory. First, the uncertain degree of the assessment is measured by the ambiguity measure. Then, the uncertainty is transformed to the reliability of each FMEA expert and the relative importance of each risk factor. After that, the assessments from FMEA team will be fused with a discounting‐based combination rule to address the potential conflict. Moreover, to avoid the situation that different risk priorities of failure modes may have the same ranking based on classical risk priority number method, the gray relational projection method (GRPM) is adopted for ranking risk priorities of failure modes. Finally, an application of the improved FMEA model in sheet steel production process verifies the reliability and validity of the proposed method.

中文翻译：

---

证据论中基于不确定性测度的改进失效模式与影响分析方法

模式和效果分析（FMEA）是评估不确定环境下系统或过程风险的有效工具。但是，如何处理主观评估中的不确定性是一个悬而未决的问题。本文在Dempster-Shafer证据理论的框架下，提出了一种新的方法来处理FMEA专家的主观评估带来的不确定性。首先，评估的不确定程度通过歧义度量来度量。然后，将不确定性转化为每个FMEA专家的可靠性以及每个风险因素的相对重要性。之后，FMEA团队的评估将与基于折扣的合并规则融合在一起，以解决潜在的冲突。此外，为了避免基于经典风险优先级数方法的故障模式不同风险优先级可能具有相同的排序，采用灰色关联投影法（GRPM）对故障模式风险优先级进行排序。最后，将改进的FMEA模型应用于钢板生产过程，验证了该方法的可靠性和有效性。