基于ARL的T^2控制图多元质量特性统计设计经济模型

ARL-Based Statistical Design of T^2 Control Chart for Process Monitoring and Economic Model

为了节约工业生产制造中经济成本,以平均运行长度ARL为标准在不同监控规则和不同相关联性下对多元T2控制图的监测性能进行分析比较,对成本经济模型进行优化,并结合中国某飞机制造集团公司实际生产数据进行分析。实证结果表明:在监控规则3下,用T2控制图进行统计过程监控控制,采样容量定为11,经济成本参数有效控制。本文所提统计设计经济模型是正确并有效的,并能够为中国某飞机制造公司实际生产提供指导。

In order to minimize costs in manufacturing processes, the average run length （ARL） is used as standard multivariate T2 control chart for process monitoring, performance analysis, and comparison between different monitoring rules. Then, an economic model is presented to minimize the cost. The proposed approach is applied to an industrial case study from an aircraft manufacturing company in China. Empirical results show that, by monitoring the process under Rule 3, T2 control charts for statistical process monitoring control are obtained. With the sampling capacity set at 11, the control of the cost parameters is effective. The proposed statistical design economic model is correct and effective and provides guidance to the actual production of an aircraft manufacturing company in China.