Abstract: To consider the problem of expert's preference correlation and psychological behavior in the traditional Failure Mode and Effects Analysis (FMEA),a FMEA risk assessment method based on Choquet integral and Tomada de Decisao Interativa e Multicritevio (TODIM) was proposed.Interval valued intuitionistic fuzzy numbers were used to characterize the risk assessment information of fuzziness and uncertainty.2-additive fuzzy measure and Choquet integration which could be used to describe the preference information correlation of experts was determined for weights of experts and risk factors.The expert evaluation information was aggregated with interval valued intuitionistic weighted arithmetic aggregation operator.The method of TODIM that could response psychological behavior of experts was calculated to the dominant degree of failure mode,and the risk ranking of failure mode was obtained.The effectiveness and feasibility of the proposed method were verified by the case of failure risk assessment of metro vehicle door system.

Key words: subway vehicle door system, risk assessment, failure mode and effects analysis, Choquet integral, tomada de decisao interativa e multicritevio, fault diagnosis

摘要： 为了在传统失效模式及影响分析(FMEA)中考虑专家偏好信息关联与心理行为,提出基于Choquet积分与交互式多准则决策(TODIM)的FMEA风险评估方法。该方法采用区间直觉模糊数表征风险评估信息的模糊性与不确定性,利用能够刻画专家偏好信息关联的2-可加模糊测度和Choquet积分确定专家和风险因子权重,运用区间直觉模糊加权算术集结算子集结专家评价信息,应用反映专家风险评估中心理行为的TODIM方法确定失效模式占优度,以此获取失效模式风险排序。通过地铁车门系统失效风险评估案例验证了所提方法的有效性与可行性。

关键词: 地铁车门系统, 风险评估, 失效模式及影响分析, Choquet积分, 交互式多准则决策, 故障诊断