Dempster-Shafer Theory Approach to FMEA
Dempster-Shafer Theory Approach to FMEA
复制标题
FMEA 的 Dempster-Shafer 理论方法
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Aric R. Johnson
中科院分区：
文献类型：
--
作者：
Niranjan S. Kulkarni;Aric R. Johnson
失效模式和影响分析（FMEA）技术是一种基于组的风险排序方法。通常情况下，汇总团队成员的反应或评级会出现问题。文献提出了几种汇总这些回答的方法：计算平均值，使用群体共识，为团队成员的回答分配权重，以及使用模糊逻辑/推理系统，等等。然而，每种方法都有一定的缺点。此外，面对不确定性或不完整的信息，对可能的失效模式进行排序也有其自身的挑战。前面提到的聚合技术在这种不确定的情况下使用的效果非常有限。本文将提出一种利用Dempster$Shafer理论（DST）来克服这些问题的方法。它将显示，DST将允许使用证据组合规则的团队成员的评级的无偏聚合。此外，这种方法将有助于在存在不完整信息的情况下精确评估因素的困难。据我们所知，在文献中，基于DST的技术尚未应用于FMEA。本文的第二个贡献是介绍了DST的概念，它在工业和系统工程领域有许多潜在的应用。
Failure Mode and Effects Analysis (FMEA) technique is a group based activity to prioritize risk. Often times there are issues with aggregating team member’s responses or ratings. Literature suggests several means to aggregate these responses $ computing average values, using g roup consensus, assigning weights to team member’s responses, and using fuzzy logic / inference systems, to name a few. However, each of these methods has certain shortcomings. Furthermore, ranking the possible failure modes in face of uncertainty or incomplete information presents its own challenges. The aforementioned aggregation techniques have been used with very limited success in such uncertain instances. This paper will propose a method that utilizes Dempster$Shafer Theory (DST) to overcome the se issues. It will be shown that DST will allow unbiased aggregation of team member’s ratings using rules of combination of evidence. Also this method will aid the difficulty in precisely rating a factor in the presence of incomplete information. To the best of our knowledge a DST based technique has not been applied to FMEA in the literature. A secondary contribution of this paper is introducing the concept of DST, which has many potential applications in the field of Industrial and System Engineering.