Abstract: For randomness and fuzziness of the two uncertain factors in the process of risk assessment, cloud model is used to quantize the linguistic evaluation variable, and the synthesis of cloud model is used to integrate the evaluation of experts, then it establishes the cloud risk assessment matrix. The weights of risk factors that O、S、D are calculated by combing expectations importance and similarity measurement. Considering the defect of value-at-risk ranking in traditional FMEA, the cloud-VIKOR method is proposed to rank the value-at-risk of failure mode based on the distance measure about the cloud models. Finally, an example is given out to demonstrate the accuracy and effectiveness of the proposed approach.

Key words: Failure Mode and Effects Analysis（FMEA）, cloud model, similarity, distance measure, Vlsekriterijumska Optimizacija I Kompromino Resenje（VIKOR）

摘要： 针对风险评估过程中的模糊性和随机性两个不确定性因素，采用云模型对语义评价变量进行量化，并通过云的合成对专家的评价结果进行集结，构建云风险评价矩阵。将风险因子O、S、D期望重要度与云的相似度相结合计算风险因子的权重。考虑到传统故障模式与影响分析（Failure Mode and Effects Analysis，FMEA）中故障模式风险值排序的缺陷，基于云的距离测度算法提出云-VIKOR的风险排序方法。最后以实例验证了所提方法的精确性和有效性。

关键词: 故障模式与影响分析（FMEA）, 云模型, 相似度, 距离测度, 模糊多准则妥协解排序法（VIKOR）