Faculty Publications

A Multivariate Methodology For Controlling Industrial Processes

Document Type

Article

Journal/Book/Conference Title

IEEE Transactions on Industrial Electronics and Control Instrumentation

Volume

IECI-22

Issue

2

First Page

175

Last Page

178

Abstract

The combination of three multivariate techniques (Principal Component Analysis, Cluster Analysis and Discriminant Analysis) may be used to isolate and determine bounds for the critical-control variable in a “black box” type industrial process. If input samples can be grouped into homogeneous quality categories, then it may be possible to use input and operating data to estimate discriminant functions which will differentiate between various quality categories These functions could then be used to predict quality levels based upon input and operating characteristics. In addition, -tne coefficients of the discriminant functions would provide informa-Hon as to the critical control variables and″ their bounds. © 1975, IEEE. All rights reserved.

Department

Department of Business

Original Publication Date

1-1-1975

DOI of published version

10.1109/TIECI.1975.351248

Link to Full Text

Comments?

Share

COinS