Gaussian process modeling and optimization of profile response experiments

Hussam Alshraideh, Enrique Del Castillo

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Experiments where the response of interest is a curve or 'profile' arise in a variety of applications in engineering practice. In a recent paper (Journal of Quality Technology, 44, 2, pp. 117-135, 2012), a mixed-effects Bayesian approach was proposed for the Bayesian optimization of profile response systems, where a particular shape of the profile response defines desired properties of the product or process. This paper proposes an alternative spatio-temporal Gaussian random function process model for such profile response systems, which is more flexible with respect to the types of desired profile shapes that can be modeled and allows us to model profile-to-profile correlation, if this exists. The method is illustrated with real examples taken from the literature, and practical aspects related to model building and diagnostics are discussed.

Original languageEnglish (US)
Pages (from-to)449-462
Number of pages14
JournalQuality and Reliability Engineering International
Volume30
Issue number4
DOIs
StatePublished - Jun 2014

Fingerprint

Experiments
Gaussian process
Process modeling
Experiment
Process optimization
Bayesian approach
Diagnostics
Process model

All Science Journal Classification (ASJC) codes

  • Safety, Risk, Reliability and Quality
  • Management Science and Operations Research

Cite this

@article{387597f41c5744078008f1a12cf0cca6,
title = "Gaussian process modeling and optimization of profile response experiments",
abstract = "Experiments where the response of interest is a curve or 'profile' arise in a variety of applications in engineering practice. In a recent paper (Journal of Quality Technology, 44, 2, pp. 117-135, 2012), a mixed-effects Bayesian approach was proposed for the Bayesian optimization of profile response systems, where a particular shape of the profile response defines desired properties of the product or process. This paper proposes an alternative spatio-temporal Gaussian random function process model for such profile response systems, which is more flexible with respect to the types of desired profile shapes that can be modeled and allows us to model profile-to-profile correlation, if this exists. The method is illustrated with real examples taken from the literature, and practical aspects related to model building and diagnostics are discussed.",
author = "Hussam Alshraideh and {Del Castillo}, Enrique",
year = "2014",
month = "6",
doi = "10.1002/qre.1497",
language = "English (US)",
volume = "30",
pages = "449--462",
journal = "Quality and Reliability Engineering International",
issn = "0748-8017",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

Gaussian process modeling and optimization of profile response experiments. / Alshraideh, Hussam; Del Castillo, Enrique.

In: Quality and Reliability Engineering International, Vol. 30, No. 4, 06.2014, p. 449-462.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Gaussian process modeling and optimization of profile response experiments

AU - Alshraideh, Hussam

AU - Del Castillo, Enrique

PY - 2014/6

Y1 - 2014/6

N2 - Experiments where the response of interest is a curve or 'profile' arise in a variety of applications in engineering practice. In a recent paper (Journal of Quality Technology, 44, 2, pp. 117-135, 2012), a mixed-effects Bayesian approach was proposed for the Bayesian optimization of profile response systems, where a particular shape of the profile response defines desired properties of the product or process. This paper proposes an alternative spatio-temporal Gaussian random function process model for such profile response systems, which is more flexible with respect to the types of desired profile shapes that can be modeled and allows us to model profile-to-profile correlation, if this exists. The method is illustrated with real examples taken from the literature, and practical aspects related to model building and diagnostics are discussed.

AB - Experiments where the response of interest is a curve or 'profile' arise in a variety of applications in engineering practice. In a recent paper (Journal of Quality Technology, 44, 2, pp. 117-135, 2012), a mixed-effects Bayesian approach was proposed for the Bayesian optimization of profile response systems, where a particular shape of the profile response defines desired properties of the product or process. This paper proposes an alternative spatio-temporal Gaussian random function process model for such profile response systems, which is more flexible with respect to the types of desired profile shapes that can be modeled and allows us to model profile-to-profile correlation, if this exists. The method is illustrated with real examples taken from the literature, and practical aspects related to model building and diagnostics are discussed.

UR - http://www.scopus.com/inward/record.url?scp=84901469230&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901469230&partnerID=8YFLogxK

U2 - 10.1002/qre.1497

DO - 10.1002/qre.1497

M3 - Article

AN - SCOPUS:84901469230

VL - 30

SP - 449

EP - 462

JO - Quality and Reliability Engineering International

JF - Quality and Reliability Engineering International

SN - 0748-8017

IS - 4

ER -