小批量生产的单值多变量控制图

Multivariate Control Charts for Individual Observations in Low-volume Production

构制单值多变量控制图的一般方法是根据T 2 统计量。常规的T 2 控制图在设置控制限时要求充分多的样本数据。对小批量生产的过程或在生产开始阶段 ,常规的T 2 图就不能有效地使用。文章研究样本数据较少时T 2 控制图的制订方法 ,使得不论抽取多少数据 ,都能可靠地使用 ,并对T 2 统计量超出控制限的原因作出解释。

Conventional Control limits for multivariate control charts should be based on a large number of data,so that it is not possible to apply control charts in Low-volume production.This paper presents control limits for multivariate control charts based on a smaller number of data,they can be reliably used regardless of how few product have been inspected.Moreover,discusses applying T 2 decomposition and control ellipse to interprete the cause of signals given by a T 2 statistic in a multivariate control chart.