The MVPMODEL, MVPMONITOR, and MVPDIAGNOSE procedures, referred to collectively as the MVP procedures, are used together for multivariate process monitoring. The MVPMODEL and MVPMONITOR procedures were introduced as experimental procedures in SAS/QC 9.3 and are production for SAS/QC 12.1. The MVPDIAGNOSE procedure is new in SAS/QC 12.1.

The MVPMODEL procedure provides computational and graphical tools for building a principal component model from multivariate process data in which the measured variables are continuous and correlated. It implements principal component analysis (PCA) techniques that evolved in the field of chemometrics for monitoring hundreds or even thousands of correlated process variables; see Kourti and MacGregor (1995, 1996) for an introduction. A principal component model reduces the dimensionality of the data by projecting the process measurements to a low-dimensional subspace that is defined by a small number of principal components. This subspace is known as the model hyperplane.

The principal component model and other output from PROC MVPMODEL serve as input to the MVPMONITOR and MVPDIAGNOSE procedures.

The MVPMONITOR procedure creates control charts for multivariate process data by using the principal component model that is produced by the MVPMODEL procedure. Multivariate control charts detect unusual variation that would not be uncovered by individually monitoring the process variables with univariate control charts, such as Shewhart charts. PROC MVPMONITOR creates two types of multivariate control chart. charts detect unusual variation within the model hyperplane, whereas squared prediction error (SPE) charts detect unusual variation from the hyperplane.

The MVPDIAGNOSE procedure produces principal component score plots and process variable contribution plots that are used to investigate the causes of unusual variation in a process.

See Chapter 10: Introduction to Multivariate Process Monitoring Procedures, for a thorough description of how the MVP procedures work together.