Research On PFMEA Risk Assessment Theory For Small-batch Customization Production And Its Application

With the development of the economy,the consumers’ demand for personalized customization products is increasing and stimulate more and more enterprises to adopt small-batch customization production mode.In order to ensure the quality of small-batch customized products,it is necessary to effectively control and continuously improve the quality of the corresponding manufacturing process,and it is urgent to carry out the research on the theory for quality management of small-batch customization production process and its engineering application.Process Failure Mode and Effects Analysis(PFMEA)is a forward-looking reliability analysis technology dedicated to improving product quality.PFMEA risk assessment is an important part of PFMEA technology and provides a basis for determining the key process failure modes by scientifically evaluating the risk level of each process failure mode.However,the existing PFMEA risk assessment method based on huge amounts of data accumulated in mass production process is not suitable for small-batch customization production process in which the components of production process are frequently adjusted and the related process quality data is difficult to accumulate.Hence,PFMEA risk assessment for small-batch customization production is taken as the research object in this paper.The expression of risk assessment opinion,the aggregation method of experts’ opinions,the aggregation method of assessment opinions for risk factors,and so on are studied to build systemic and scientific theory of process failure mode risk assessment.The main research work is as follows:Firstly,the risk assessment opinion expression is studied.The on-site situation of small-batch customization production is complex and variable,which requires a variety of risk assessment opinions.Hence,the risk assessment opinion expression based on the improved model of the assessment hypothesis is proposed,which require the experts to report the possible intervals containing the true value of the risk factor,the data distribution status within each possible interval,and the level of credibility for each evaluation conclusion.These three aspects form the complete risk asssessment opinion which is converted into the corresponding confidence fuzzy interval number by the fuzzy factors.Secondly,the PFMEA risk assessment problem based on team consensus is studied.The PFMEA risk assessment opinion for the risk factor is uniformly given after sufficient discussion among PFMEA team members.In order to obtain the comprehensive risk level of the failure mode,it is necessary to aggregate the assessment opinions of risk factors.In addition,one of main features of small-batch customization production model is frequent adjustments of product types,which leads to frequent adjustment of risk factor weights.To flexibly adjust risk factor weights and effectively process confidence fuzzy interval number,a hybrid method drawn from Hausdorff-distance-based Grey Relational Analysis-Technique for Order Preference by Similarity to an Ideal Solution(HGRA-TOPSIS)is proposed.Its characteristics is that the risk assessment opinion is regarded as a random variable,and the corresponding probability density function is constructed to calculate generalized Hausdorff distance between confidence fuzzy interval numbers.Case studies show that the method has high stability and feasiblity.Thirdly,the PFMEA risk assessment problem based on the aggregation of experts’ opinions is studied.Faced with the complex risk situation of small-batch customization production process,it is often difficult for experts to reach a completely consistent opinion,which requires the appropriate method to aggregate different expert opinions into group expert opinion.Hence,a hybrid method drawn from Hausdorff-distance-based and Matrix-sequence-based Rough Grey Relational Analysis-Technique for Order Preference by Similarity to an Ideal Solution(HMRGRA-TOPSIS)is proposed.Case study shows that HMRGRA-TOPSIS method can effectively simulate the actual process of forming consensus in the meeting and achieve reasonable risk prioritization.Fourthly,evaluation method of risk factors is studied.The lack of relevant data makes PFMEA risk assessment for small-batch customization production difficult and traditional evaluation method of risk factor error-prone.Hence,a staged and multi-dimensional evaluation method of risk factors based on fuzzy evidence theory is proposed.The method can help experts to concentrate on in-depth vertical thinking,and effectively prevent the omission of the evaluation dimension.In addition,systematic theories are used to construct and optimize the multi-dimensional evaluation criteria system of severity to ensure the effectiveness and scientificity of risk assessment.Fifthly,the research results are summarized to form PFMEA risk assessment theory for small-batch customization production,which is applied to aircraft canopy assembly process.the validity of the theoretical result is verified by its comprehensive application.The PFMEA risk assessment problem of small-batch customization production is systematically studied in this paper.The problems constraining the risk assessment have been innovatively solved from the aspects of risk assessment opinion expression,the aggregation of evaluation opinions,multi-dimensional evaluation of risk factor,and so on.The systemic and scientific PFMEA risk assessment theory for small-batch customization production has been formed and can provide scientific decision-making basis for continuously controlling and improving the quality of small-batch customization production process.