基于PFMEA技术及专家群决策的工艺失效风险评估

摘要

The traditional process failure risk assessment is usually based on the limited experience accumulation of experts or similar products manufacturing process.The assessment process has some complexity and cognitive uncertainty.In order to evaluate the process objectively and accurately model the risk of failure, this paper proposed a failure risk assessment process and PFMEA technology based on expert group decision making method.Based on the traditional PFMEA analysis techniques, the team of experts gave the assessment opinion of each process failure mode under the three risk factors of severity, occurrence probability and detectability based on the experience and historical data with linguistic variables.The Euclidean distance between trigonometric fuzzy numbers was used to calculate the similarity of evaluation opinions among different experts so as to obtain the experts'weights under different risk factors and then to obtain the expert opinion after assembly.According to the fuzzy synthesis rule in fuzzy theory,the fuzzy risk priority numbers of failure modes could be obtained,and the exact RPN values could be obtained by defuzzification,and then the risks of technological failure modes could be sorted according to the precise risk priority numbers.The numerical example showed the concrete realization process of this method and verified its validity and feasibility.%传统工艺失效风险评估通常依据专家或类似产品制造过程的有限经验积累来识别,评估过程具有一定复杂性、认知不确定性.为客观准确地评估工艺失效模式的风险,提出一种基于PFMEA(Process Failure Mode and Effects Analysis)技术和专家群决策的工艺失效风险评估方法.专家团队根据经验和历史数据在传统PFMEA分析技术基础上,以语言变量给出对严酷度、发生概率和可探测度三个风险因子下各工艺失效模式的评估意见,采用三角模糊区间数的欧式距离计算各专家评估意见相似度,由此得到不同风险因子下的专家权重,然后求取集结后的专家评估意见;根据模糊理论中的模糊合成法则,可求取失效模式的模糊风险优先数,通过去模糊化得到精确RPN值,根据精确风险优先数来进行工艺失效模式风险排序.数值实例说明了该方法的具体实现过程,并验证了其有效性和可行性.