可变抽样区间和样本容量的非正态EWMA控制图经济设计

为了降低过程控制成本和提高监控效率,针对质量特性值不服从正态分布的情况,研究可变抽样区间和样本容量(VSSI)的指数加权移动平均(EWMA)控制图的经济设计问题。首先对监控非正态分布的EWMA控制图进行抽样区间和样本容量变化设计;其次建立VSSI 非正态EWMA控制图的经济模型,通过使费用成本函数最小得到控制图的最优设计参数组合;然后给出工业中的一个例子,用遗传算法对经济模型搜寻最优解;接下来对经济模型进行灵敏度分析,得出控制图费用参数与设计参数之间的影响关系;最后通过最优性分析,得出所建立的VSSI 非正态EWMA控制图的经济性优于VSI、VSS非正态EWMA控制图。

Economic Design of Variable Sampling Interval and Sample Size EWMA Control Chart under Non-normal Distributions

To reduce the costs of process control and improve the monitoring efficiency of control chart,the economic design of exponential weighted moving average(EWMA)control chart with variable sampling intervals and sample size(VSSI)for monitoring non-normal data is proposed.Firstly,using Burr distribution to approximate various non-normal distributions,EWMA control chart with variable sampling intervals and sample size is proposed.Secondly,economic model of VSSI EWMA control chart under non-normal distributions was developed,the optimal design parameter combination of the control chart is obtained,such that the total expected cost is minimized.Thirdly,the genetic algorithms(GA)will be used to search for the optimal values of parameters of the VSSI EWMA chart for non-normal data,and an example is provided to illustrate influence relationship between the cost parameters of the control chart and the design parameters.Finally,through the optimality analysis,it is concluded that the economy of the established VSS Inon-normal EWMA control chart is better than that of VSI and VSS non-normal EWMA control chart.