蒙特卡罗方法的过程聚类EWMA控制图研究

Study on EWMA control chart for the process clustering based on monte carlo method

以加工"过程"为研究对象,从零件成组加工技术出发,阐述了就如何利用EWMA控制图来进行统计聚类并进行过程质量监控等问题,提出通过统计聚类形成"零件族",对同一"零件族"采用统一的EWMA控制图的应用策略;引入聚类零件族的方差比ρ,利用蒙特卡罗模拟方法,对不同样本数量条件下的统计聚类EWMA控制图性能所产生的影响进行了详细的研究,最后给出了统计聚类的基本原则.

The machining process was studied. From the view of group machining technology, it discussed how to use the EWMA control chart to achieve the purpose of statistical clustering and then used the result of statistical cluster to monitor the quality of machining processes. The method of using statistical clustering to attain Part Family was carried out. The unified EWMA control chart was used on the same Part Family. The pa rameter of the ratio of clustered Part Family Variables was introduced. Using the method of Monte Carlo simulation, a detailed analysis of th EWMA control chart for process clustering was given under different sample numbers. At last, the basic principles of using statistical clustering are proposed.