Abstract Failure mode effects and criticality analysis (FMECA) is widely used, by developing a Risk Priority Number (RPN), to identify the failure modes and to prioritize them. But this has been extensively criticized due to several drawbacks in the literature. This issue can be solved partly by using Fuzzy FMECA (FFMECA) although Fuzzy logic itself has been criticized of having a direct bearing on subjectivity. This paper makes use of Dempster-Shafer Theory (DST) of evidence-a proper mathematical framework to deal with the epistemic uncertainty often affecting the input evaluations of risk parameters. DST based FMECA is capable of providing an appropriate, precise and fault-free, failure mode prioritization. Belief and Plausibility distributions are used to synthesize the obtainable information and to make them useful for the purpose. The results obtained from DST-FMECA is compared to the results drawn from the FFMECA applications (already conducted in the liquefied natural gas (LNG) storage facility) to validate FFMECA and vice versa. The comparative results presented in this paper establish the capabilities of both the approaches, especially in a complex system like the LNG storage and similar facilities where even a minor failure may lead to catastrophic effects.

中文翻译：

---

使用 Dempster Shafer 理论的失效模式影响和临界分析及其与模糊失效模式效应和临界分析的比较：应用于 LNG 储存设施的案例研究

摘要 故障模式影响和关键性分析 (FMECA) 被广泛使用，通过开发风险优先级数 (RPN) 来识别故障模式并对其进行优先级排序。但由于文献中的几个缺点，这已受到广泛批评。这个问题可以通过使用模糊 FMECA (FFMECA) 部分解决，尽管模糊逻辑本身被批评为与主观性有直接关系。本文利用Dempster-Shafer Theory (DST) of evidence——一个合适的数学框架来处理经常影响风险参数输入评估的认知不确定性。基于 DST 的 FMECA 能够提供适当、精确和无故障的故障模式优先级。置信度和合理性分布用于综合可获得的信息并使它们对目的有用。将从 DST-FMECA 获得的结果与从 FFMECA 应用（已在液化天然气 (LNG) 储存设施中进行）得出的结果进行比较，以验证 FFMECA，反之亦然。本文中提供的比较结果证明了这两种方法的能力，尤其是在像 LNG 储存和类似设施这样的复杂系统中，即使是轻微的故障也可能导致灾难性的影响。