Statistical process and controller performance monitoring. A tutorial on current methods and future directions

摘要

Statistical process control procedures for a single variable such as Shewhart, CUSUM and EWMA charts are summarized. Multivariable statistical process monitoring (SPM) techniques for both continuous and batch processes are presented. SPM methods based on principal components analysis, canonical variate state space models, partial least squares, multiway and multiblock techniques are discussed. Methods for monitoring single loop and multivariable controller performance are also presented.