单边MOEWMA CV控制图及其在合金烧结工艺中的应用OA

In some financial,textile and industrial processes,when the process mean μ or standard devia-tion σ changes and there is a fixed proportional relationship,the process is still in-control.In these cases,it is difficult to use the traditional mean(Xˉ)or variance(S2)chart for the effective monitoring of the pro-cesses.Using control charts to monitor process Coefficient of Variation(CV)can solve this problem effec-tively.In order to improve the performance of existing Exponentially Weighted Moving Average(EWMA)CV control chart,we considered the sample CV fluctuations at different times,and used coeffi-cients to control the weight of the fluctuations in the monitoring statistic.A one-sided Modified EWMA(MOEWMA)control chart was proposed to monitor CV.Due to the complexity of the proposed monitor-ing statistic,based on 104 times simulations,this paper used Monte Carlo method to simulate the Run Length(RL).For different parameter combinations,the corresponding control limits were calculated by using the bisection algorithm.The optimal parameter combinations(λ,K)and the corresponding ARL1 of the control chart were designed by satisfying the desired ARL0 for different parameter changes.Based on the designed parameters,the ARL performance of MOEWMA CV control chart was simulated and com-pared with the EWMA CV chart.The simulation results show that the MOEWMA CV is superior to the EWMA CV control chart for downward changes(τ∈{0.5,0.65,0.8,0.9})and upward small(τ=1.1)and medium(τ∈{1.25,1.5})changes.Finally,the performance advantage of the proposed MOEWMA CV chart was verified in monitoring the actual alloy sintering process.