高质量过程控制技术的比较研究与分析

Comparison and Analysis for High Quality Processes Monitoring Techniques

随着质量改进活动的不断开展,现代制造过程中的不合格项在逐渐降低。在这种情况下,常规的休哈特型计数控制图往往是失效的。为了监控高质量的过程运行,一种方法是采用累计合格品计数(CCC)图;另一种方法是采用几何Q图,本文首先分析了这两种控制图的基本原理;进而以平均链长(ARL)和探测过程发生漂移的概率为准则,系统分析和比较了这两种控制图的性能,仿真结果表明,在大多数情况下,这两种控制图具有相似的性能;最后,通过实例说明了这两种控制图的应用,并给出了若干建议。

With the development of quality improvement programs, many production processes today .are producing a very low level of nonconforming items. In this case, conventional Shewhart type attribute control charts are not applied effectively in practice. In order to monitor high quality processes, two alternative techniques are proposed： one is a cumulative count of conformance （CCC）chart, the other is geometric Q chart. In this paper, we first analyze basic principal of both CCC chart and geometric Q chart, which is based on the geometric distribution. Then we compare the performance by using average run length （ARL） and the probability of detecting a shift in the observations after a shift occurs. Simulation results show that in most cases CCC chart and geometric Q chart are of the similar performance. After that an example from practice is presented. And finally, some suggestions are given in conclusion.