The Research On The Adaptive Control Chart In Statistical Process Control

The competition between the products boils down to product quality,because the quality is the core competitiveness of the products.Since the 21st century is the century of quality,how to improve the product quality becomes a focus of research.Control chart,as an important tool of statistical quality control,can effectively ensure the quality of products conforming to requirements through online monitoring.The problems of insensitivity to small shifts of Shewhart control chart and the inertia problem of EWMA control chart have long been the key research area of control chart.To solve this problem,the early method of adding Shewhart limits into EWMA control chart directly has been gradually replaced by more effective one.The Adaptive EWMA control chart formed by the combination of Shewhart and EWMA control chart by means of adaptive EWMA filtering method has a more superior performance and greater research value.In addition,more and more manufacturers tend to use small batch production in order to better adapt to the needs of market competition.The application of traditional control charts in a small batch process has been limited,because of the establishment of traditional control chart needs infinite sample space.To solve this problem,scholars have proposed the new statistics.In recent years,the t control chart based on T statistic has drawn wide attention due to its excellent performance.However,the overall performances of Shewhart t and EWMA t control chart is limited by the shortcomings of Shewhart and EWMA control chart.Due to the superior performances of AEWMA and t control chart,the AEWMA t control chart formed by the combination of them is worthy of study.Based on this consideration,the performance of AEWMA t control chart is studied in this paper.The main research results of this dissertation are summarized as follow:1.The statistical design of AEWMA t control charts with unknown variance is investigated.Firstly,the performance indexes average run length(ARL)and standard deviation of run length(SDRL)of AEWMA t control chart are derived by the Markov chain method.Considering the fact that the process variance is unknown in the actual production process,the variance will be inaccurate even if it can be estimated by statistical method.The performance of AEWMA t control chart is investigated,and the influence of error size on the statistical performance of AEWMA t control chart is analyzed.The statistical characteris-tics of AEWMA t and EWMA t control charts are compared.The effectiveness of AEWMA t control chart in solving the inertia problem of the EWMA t control chart has been verified.2.The economic statistical design method of AEWMA t control chart is investi-gated.The traditional economic design will reduce statistical performances of control chart,so the economic statistical design is a more comprehensive effective design method.Two kinds of economic statistical design models are proposed,based on the thoroughly investigating of economic statistical design method of AEWMA t control chart.The economic and statistical properties of the AEW-MA t control chart under two models are analyzed.And the sensitivity analysis of the key parameters of the AEWMA t control chart is computed.3.The statistical design of AEWMA t control chart in short run production is investigated.The performance indexs truncated ARL(TARL)and alarm prob-ability q of the short run AEWMA t control chart are derived by the Maxkov chain method.The statistical design of AEWMA t control chart is carried out based on the TARL index.The statistical performances of short run AEWMA t control chart and EWMA t control chart are compared.And the performance of short run AEWMA t control chart in solving the inertia problem is investigated.4.The economic design of the short run AEWMA t control chart are investigated.The Taguchi loss function is introduced into the existing economic design model,which makes the model more conforming to reality.Two dimensional Maxkov chain method for calculating the loss is derived.The economic performances of the short run AEWMA t control chart are analyzed,the results show that the cost of the AEWMA t control chart is less than the EWMA t control chart and more robust than AEWMA X control chart to variance uncertainty.The sensitivity analysis of the key parameters is computed.The effects of the parameters on the decision variables and the loss of the short run AEWMA t control chart are analyzed quantitatively.