|
Abstract:
|
Monitoring of processes that involve multiple variables through control charts traditionally have used a Mahalanobis distance measure in the form of a T-squared control chart. The variables used normally have been the process output variables. However, the process output variables could be impacted by two categories of variables: process regulatory or input variables and the uncontrollable noise factors. While the process input variables are usually assumed to be fixed, this is not the case in practice. Hence, we propose the inclusion of these process input variables in the analysis of multivariate process control. For improved detection of out-of-control conditions, it is of interest to determine which of these categories of variables are potential causes. Through a simulation study, certain performance measures are investigated.
|
