MONITORING A MULTIVARIATE STEP PROCESS

The productivity of an industrial processing unit often depends on equ ipment that changes over time. These changes may not be consistent, an d, in many cases, may appear to occur in stages. Although changes in t he process levels within each stage may appear insignificant, they can be substantial when monitored across the various stages. Standard pro cess control procedures do not perform well in the presence of these s tep-like changes, especially when the observations from stage to stage are correlated. This paper presents an alternative multivariate contr ol procedure for monitoring a process under these conditions. The meth odology is based on a double decomposition of Hotelling's T-2 statisti c.