Risk Assessment Of Failure Mode Based On Utility Function And Fuzzy Reasoning

Failure mode and effect analysis is a widely used reliability analysis method and management tool,and risk assessment of failure mode is its key content.In view of the shortcomings of the fifth edition of the failure mode and effect analysis standard widely used in the industry,such as not reflecting the impact of the objectively existing evaluator's risk preference and the possibility of serious risk tie so that evaluator can't distinguish key failure mode effectively,the following research work of risk assessment of failure modes in this thesis is carried out systematically by using utility function and fuzzy reasoning theory.Firstly,the risk analysis of failure modes is carried out.The risk matrix method is used to construct a three-dimensional risk matrix from three evaluation dimensions of severity,occurrence and detection.Then,the fuzzy set theory is introduced to quantitatively calculate the risk factors,and the risk factor level of the failure modes is obtained.Secondly,according to the constructed three-dimensional risk matrix,the two-dimensional risk index under different risk factor levels is defined,and then the risk preference differential model based on the exponential utility function is established,the final three-dimensional risk matrix grade distribution was determined by parameter selection,and the risk matrix chromatogram and risk radar chart are used to represent the risk grade distribution.Third,the fuzzy logic inference system is used to fuzzify the risk factors and their risk levels,and the fuzzy rules in the fuzzy logic inference system are established according to the risk level category distribution obtained by the exponential utility function to obtain the severity,frequency,detection and the mapping relationship between risk levels,the risk value is obtained by the center of gravity defuzzification method.Finally,the risk factor level of each failure mode is combined with the risk matrix level distribution to determine the final failure mode risk level,and the failure modes with high risk levels are selected for further differentiation.By applying the quantitative evaluation results of risk factors to the fuzzy logic inference system,the risk value of each failure mode is finally obtained,and the evaluation results of different methods are compared and analyzed to verify the effectiveness of the given model.The research results of the thesis show that the proposed risk assessment model of failure mode is more precise and stable in distinguishing the degree of risk of failure modes than the traditional assessment model of risk priority number and the priority method of measures given in the fifth edition of the failure mode and effect analysis standard.The model proposed in the thesis can effectively guide enterprises to carry out risk assessment work more scientifically.