The Economic Design Of EWMA Control Chart For Autocorrelation Process

In the process of modern manufacture,the quality control of autocorrelation process is a hot topic in the field of quality control at present.Because of the continuous improvement of production technology and the application of sensors in automatic control,the observational values of the process are autocorrelated,traditional control chart will lose effectiveness while monitoring autocorrelation process.In the design time,traditional control chart only considered its statistical characteristics and ignored the economic characteristics,therefore,this paper will study the economic design of control charts for autocorrelation processes.This paper is based on the theory of control chart,according to the economic model of the control chart proposed by Lorenzen and Vance,Introduced Taguchi's quality loss function while researching the economic design and establish the quality cost function of EWMA control chart and EWMA control chart of the autocorrelation process,using the method of Markov chain to simulate the average run length(ARL)of control charts.Then we use the genetic algorithm toolbox in matlab to solve the cost model,in the case of minimum unit time cost to solving the best combination of parameters.Finally,we used experimental design and regression analysis to analysis the sensitivity of control chart parameters,to exploring the influence of control chart model parameters on the economic effect of control charts.The innovation points are: Introduced Taguchi's quality loss function to establish the quality cost function of EWMA control chart and EWMA control chart of the autocorrelation process;Using the Markov chain to calculate ARL;Genetic algorithm is used to determine the best parameter combination of cost function.