Mean Shifts Identification in Multivariate Autocorrelated Processes Based on PSO-SVM Pattern Recognizer

Chi Zhang () and Zhen He
Additional contact information
Chi Zhang: University of Tianjin
Zhen He: University of Tianjin

Chapter Chapter 23 in Proceedings of 2012 3rd International Asia Conference on Industrial Engineering and Management Innovation (IEMI2012), 2013, pp 225-232 from Springer

Abstract: Abstract In multivariate statistical process control, interpretation of a signal issued by multivariate control charts is very useful to find source(s) of variation that result in the out-of-control condition. This paper develops a support vector machine(SVM) based model for multivariate autocorrelated processes to diagnose abnormal patterns of process mean changes, and to help identify abnormal variable(s) when residual T2 control chart issue an alarm. Particle swarm optimization (PSO) method is adopted to determine the values of penalty parameter and kernel parameter of the model to improve the performance of the SVM pattern recognizer. The results demonstrate that the proposed method provides an excellent performance in terms of accuracy of classifying patterns of out-of-control signals.

Keywords: Multivariate autocorrelated processes; Support vector machine; Particle swarm optimization; Quality diagnosis (search for similar items in EconPapers)
Date: 2013
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-642-33012-4_23

Ordering information: This item can be ordered from
http://www.springer.com/9783642330124

DOI: 10.1007/978-3-642-33012-4_23

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().