Multivariate statistical process control with artificial contrasts

Wookyeon Hwang, George Runger, Eugene Tuv

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

A multivariate control region can be considered to be a pattern that represents the normal operating conditions of a process. Reference data can then be generated and used to learn the difference between this region and random noise. Then multivariate statistical process control can be converted to a supervised learning task. This can dramatically reshape the control region and open the control problem to a rich collection of supervised learning tools. Such tools provide generalization error estimates that can be used to specify error rates. The effectiveness of such an approach is shown here. Such a computational approach is now easily accomplished with modern computing resources. Examples use random forests and a regularized least squares classifier as the learners.

Original languageEnglish (US)
Pages (from-to)659-669
Number of pages11
JournalIIE Transactions (Institute of Industrial Engineers)
Volume39
Issue number6
DOIs
StatePublished - Jun 2007

Fingerprint

Statistical process control
Supervised learning
Classifiers
Multivariate statistical process control

Keywords

  • Classification
  • Control chart
  • False alarm
  • Random forest
  • Regularization
  • Supervised learning

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Management Science and Operations Research

Cite this

Multivariate statistical process control with artificial contrasts. / Hwang, Wookyeon; Runger, George; Tuv, Eugene.

In: IIE Transactions (Institute of Industrial Engineers), Vol. 39, No. 6, 06.2007, p. 659-669.

Research output: Contribution to journalArticle

@article{f28bc6d62f874774851fea227fb47d44,
title = "Multivariate statistical process control with artificial contrasts",
abstract = "A multivariate control region can be considered to be a pattern that represents the normal operating conditions of a process. Reference data can then be generated and used to learn the difference between this region and random noise. Then multivariate statistical process control can be converted to a supervised learning task. This can dramatically reshape the control region and open the control problem to a rich collection of supervised learning tools. Such tools provide generalization error estimates that can be used to specify error rates. The effectiveness of such an approach is shown here. Such a computational approach is now easily accomplished with modern computing resources. Examples use random forests and a regularized least squares classifier as the learners.",
keywords = "Classification, Control chart, False alarm, Random forest, Regularization, Supervised learning",
author = "Wookyeon Hwang and George Runger and Eugene Tuv",
year = "2007",
month = "6",
doi = "10.1080/07408170600899615",
language = "English (US)",
volume = "39",
pages = "659--669",
journal = "IISE Transactions",
issn = "2472-5854",
publisher = "Taylor and Francis Ltd.",
number = "6",

}

TY - JOUR

T1 - Multivariate statistical process control with artificial contrasts

AU - Hwang, Wookyeon

AU - Runger, George

AU - Tuv, Eugene

PY - 2007/6

Y1 - 2007/6

N2 - A multivariate control region can be considered to be a pattern that represents the normal operating conditions of a process. Reference data can then be generated and used to learn the difference between this region and random noise. Then multivariate statistical process control can be converted to a supervised learning task. This can dramatically reshape the control region and open the control problem to a rich collection of supervised learning tools. Such tools provide generalization error estimates that can be used to specify error rates. The effectiveness of such an approach is shown here. Such a computational approach is now easily accomplished with modern computing resources. Examples use random forests and a regularized least squares classifier as the learners.

AB - A multivariate control region can be considered to be a pattern that represents the normal operating conditions of a process. Reference data can then be generated and used to learn the difference between this region and random noise. Then multivariate statistical process control can be converted to a supervised learning task. This can dramatically reshape the control region and open the control problem to a rich collection of supervised learning tools. Such tools provide generalization error estimates that can be used to specify error rates. The effectiveness of such an approach is shown here. Such a computational approach is now easily accomplished with modern computing resources. Examples use random forests and a regularized least squares classifier as the learners.

KW - Classification

KW - Control chart

KW - False alarm

KW - Random forest

KW - Regularization

KW - Supervised learning

UR - http://www.scopus.com/inward/record.url?scp=33947689147&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947689147&partnerID=8YFLogxK

U2 - 10.1080/07408170600899615

DO - 10.1080/07408170600899615

M3 - Article

AN - SCOPUS:33947689147

VL - 39

SP - 659

EP - 669

JO - IISE Transactions

JF - IISE Transactions

SN - 2472-5854

IS - 6

ER -