基于威布尔分布函数的自适应MEWMA图多目标优化设计OA北大核心CHSSCDCSSCICSTPCD

Control chart is a common statistical quality control method widely used in quality management activi-ties of various production and service processes.For the monitoring of two or more interrelated quality character-istics,the method of multivariate control charts is usually used.With the increasing complexity of production and service processes,multivariate control charts have received increasing attention in the field of quality control.In recent years,there have been many studies on improving and optimizing their monitoring efficiency.The expo-nential weighted moving average method can fully utilize current and historical sampling detection data to identify trends in process changes.Therefore,MEWMA chart constructed by combining EWMA and commonly used mul-tivariate Hotelling statistics can still be regarded as an effective multivariate quality monitoring method. In practical applications,the various control chart parameters of MEWMA remain unchanged,making it simple to apply but there is still room for further improvement in its monitoring efficiency.Therefore,this article adopts the Weibull cumulative distribution function to construct an adaptive MEWMA chart that can dynamically adjust the EWMA smoothing coefficient based on sampling data.The statistical and economic performance evaluation indicators of the control chart are relatively accurate average product length(APL)and unit product average quality cost.The calculation method of these two indicators is studied,and an adaptive MEWMA chart economic-statistical multi-objective optimization design model for a specific shift is constructed with both as objective functions.In practical applications,abnormal shifts that may occur during the process are often difficult to determine in advance,and only the approximate range of shift degree can be known.Therefore,constructing a monitoring plan for a specific shift range rather than a specific shift is more in line with practical application needs.In a certain shift range,this article first selects multiple representative specific shift values,and uses multi-objective optimization design models to calculate multiple non-inferior solutions for each specific shift value.Then,the technique for order preference by similarity to ideal solution(TOPSIS)is used to compare and rank these non-inferior solutions,thereby selecting the optimal design solution that is relatively effective for the entire shift range.The calculation steps of this optimization design method are explained through a specific example and compared with the other two adaptive multivariate control charts and existing MEWMA economic statistical design methods.The results indicate that the optimized design in this article has significant advantages in both statistical and economic performance. In this study,only adaptive adjustments are made to the smoothing coefficient,while other parameters such as sample size,sampling interval,and control limits remained fixed.The monitoring performance of MEWMA chart is significantly improved,but there is still room for further improvement.Subsequent research will consider adaptive adjustments to these control chart parameters to achieve more comprehensive adaptive dynamic optimization.