Integration of classification algorithms and control chart techniques for monitoring multivariate processes

Thuntee Sukchotrat, Seoung Bum Kim, Kwok Leung Tsui, Victoria C P Chen

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

We propose new multivariate control charts that can effectively deal with massive amounts of complex data through their integration with classification algorithms. We call the proposed control chart the 'Probability of Class (PoC) chart' because the values of PoC, obtained from classification algorithms, are used as monitoring statistics. The control limits of PoC charts are established and adjusted by the bootstrap method. Experimental results with simulated and real data showed that PoC charts outperform Hotelling's T 2 control charts. Further, a simulation study revealed that a small proportion of out-of-control observations are sufficient for PoC charts to achieve the desired performance.

Original languageEnglish
Pages (from-to)1897-1911
Number of pages15
JournalJournal of Statistical Computation and Simulation
Volume81
Issue number12
DOIs
Publication statusPublished - 2011 Dec 1

Fingerprint

Control Charts
Classification Algorithm
Chart
Monitoring
Multivariate Control Charts
Bootstrap Method
Proportion
Class
Control charts
Statistics
Simulation Study
Sufficient
Charts
Experimental Results

Keywords

  • data mining
  • Hotelling's T
  • multivariate statistical process control
  • supervised classification method

ASJC Scopus subject areas

  • Applied Mathematics
  • Statistics and Probability
  • Modelling and Simulation
  • Statistics, Probability and Uncertainty

Cite this

Integration of classification algorithms and control chart techniques for monitoring multivariate processes. / Sukchotrat, Thuntee; Kim, Seoung Bum; Tsui, Kwok Leung; Chen, Victoria C P.

In: Journal of Statistical Computation and Simulation, Vol. 81, No. 12, 01.12.2011, p. 1897-1911.

Research output: Contribution to journalArticle

@article{a0da68969b934e019548532cb1ffdf12,
title = "Integration of classification algorithms and control chart techniques for monitoring multivariate processes",
abstract = "We propose new multivariate control charts that can effectively deal with massive amounts of complex data through their integration with classification algorithms. We call the proposed control chart the 'Probability of Class (PoC) chart' because the values of PoC, obtained from classification algorithms, are used as monitoring statistics. The control limits of PoC charts are established and adjusted by the bootstrap method. Experimental results with simulated and real data showed that PoC charts outperform Hotelling's T 2 control charts. Further, a simulation study revealed that a small proportion of out-of-control observations are sufficient for PoC charts to achieve the desired performance.",
keywords = "data mining, Hotelling's T, multivariate statistical process control, supervised classification method",
author = "Thuntee Sukchotrat and Kim, {Seoung Bum} and Tsui, {Kwok Leung} and Chen, {Victoria C P}",
year = "2011",
month = "12",
day = "1",
doi = "10.1080/00949655.2010.507765",
language = "English",
volume = "81",
pages = "1897--1911",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "12",

}

TY - JOUR

T1 - Integration of classification algorithms and control chart techniques for monitoring multivariate processes

AU - Sukchotrat, Thuntee

AU - Kim, Seoung Bum

AU - Tsui, Kwok Leung

AU - Chen, Victoria C P

PY - 2011/12/1

Y1 - 2011/12/1

N2 - We propose new multivariate control charts that can effectively deal with massive amounts of complex data through their integration with classification algorithms. We call the proposed control chart the 'Probability of Class (PoC) chart' because the values of PoC, obtained from classification algorithms, are used as monitoring statistics. The control limits of PoC charts are established and adjusted by the bootstrap method. Experimental results with simulated and real data showed that PoC charts outperform Hotelling's T 2 control charts. Further, a simulation study revealed that a small proportion of out-of-control observations are sufficient for PoC charts to achieve the desired performance.

AB - We propose new multivariate control charts that can effectively deal with massive amounts of complex data through their integration with classification algorithms. We call the proposed control chart the 'Probability of Class (PoC) chart' because the values of PoC, obtained from classification algorithms, are used as monitoring statistics. The control limits of PoC charts are established and adjusted by the bootstrap method. Experimental results with simulated and real data showed that PoC charts outperform Hotelling's T 2 control charts. Further, a simulation study revealed that a small proportion of out-of-control observations are sufficient for PoC charts to achieve the desired performance.

KW - data mining

KW - Hotelling's T

KW - multivariate statistical process control

KW - supervised classification method

UR - http://www.scopus.com/inward/record.url?scp=84863230277&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863230277&partnerID=8YFLogxK

U2 - 10.1080/00949655.2010.507765

DO - 10.1080/00949655.2010.507765

M3 - Article

VL - 81

SP - 1897

EP - 1911

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 12

ER -