A PRACTICAL APPROACH FOR INTERPRETING MULTIVARIATE T-2 CONTROL CHART SIGNALS

A persistent problem in multivariate control chart procedures is the i nterpretation of a signal. Determining which variable or group of vari ables is contributing to the signal can be a difficult task for the pr actitioner. However, a procedure for decomposing the T-2 statistic int o orthogonal components greatly aids the interpretation effort. In thi s paper, a series of examples illustrates the usefulness of this inter pretation scheme and relates it to the familiar concept of a statistic al distance measure. The relationship between the T-2 decomposition pr ocedure and a regression of one variable on a subset of other variable s is also demonstrated, and used to aid signal interpretation. Recomme ndations are made and steps are suggested to provide a faster sequenti al computation scheme for the decomposition.