模型控制图是兼有一元质量控制图和多元质量控制图特点 的统计质量控制工具,其控制图的控制限取决于被控制量的概率分布,而被控 制量的概率分布则由各影响量的概率分布和模型结构所决定.文章研究了基于 蒙特卡罗(MCM)分布传播得到被控制量概率分布的相关问题,包括确定影响量 分布和MCM仿真次数等问题,给出了基于MCM 的模型质量控制图控制限计算方 法和流程,该方法与模型的复杂程度无关,与各影响量之间的相关性无关,控制图的 可靠性和监控的有效性高于传统控制图.

Model control chart is a statistical quality control tool based on the characteristics of univariate quality control chart and multivariate quality control chart. The control limits depend on the distribution of quality parameter that is determined by the distribution of influence quantities and the structure of the model. In this paper, the related issues of the spread of the Monte Carlo Method (MCM) distribution are paid attention, including identifying affect issues such as the influence quantity distribution and number of MCM simulation, and a calculation method and process of control limit of quality control chart model based on the MCM are put forward. This method does not depend on the complexity of the model and the correlation between the influence quantities. The reliability of the control chart and the effectiveness of monitor are prove to be higher than traditional control chart.