REGRESSION ADJUSTMENT FOR VARIABLES IN MULTIVARIATE QUALITY-CONTROL

Multivariate process control problems are inherently more difficult th an univariate problems. It is not always clear what type of multivaria te statistic should be used, and the most statistically powerful techn iques do not indicate the cause(s) of a signal. On the other hand, sep arate controls on the individual variables are more easily interpretab le but may be substantially less powerful, particularly in the face of appreciable correlation between the measures. Previous research has d emonstrated the effectiveness of methods that capitalize on the likely nature of a departure from control. If only one variable is likely to undergo a shift in mean or variance then charting of each variable ad justed by regression for all others is particularly effective. In this paper, the issue of other possible regression adjustments is discusse d. In particular, a regression of each variable for those driving it i s considered. It is shown under what circumstances this adjustment is appropriate, and its diagnostic power is illustrated by examples.