Approach to identifying out-of-control variables in multivariate T2 control chart using AIC

The control chart is a primary tool of judging whether a manufacturing process is in-control or not. Especially, the T2 control chart is known as a multivariate control chart that monitors a mean vector of several related quality characteristics. It is an important issue to identify which quality characteristics are responsible for out-of-control signal when a multivariate control chart signals. This paper considers a method of identifying quality characteristics responsible for out-of-control signal on operating the T2 control chart.