Bayes linear Bayes graphical models in the design of optimal test strategies

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Test and analysis plays a vital role in reducing uncertainty about the true performance of an engineering system. However tests can be expensive and designing an optimal test strategy can be challenging. We propose a Bayesian modelling process, which takes the form of a Bayesian Network, to determine anticipated test efficacy. Such a model supports engineering managers in assessing trade-offs between test resources and uncertainty reduction. Inference based on a full Bayesian model can be computationally demanding to the extent that it can limit practical application. To overcome this constraint, we develop a Bayes linear approximation for inference. This approach is known as a Bayes linear Bayes graphical model. After explaining the key principals of the method, we provide an application to a real industrial test to establish the condition of an ageing engineering system.
LanguageEnglish
Pages715-728
Number of pages14
JournalInternational Journal of Performability Engineering
Volume9
Issue number6
Publication statusPublished - Nov 2013

Fingerprint

Systems engineering
Bayesian networks
Managers
Aging of materials
Graphical models
Optimal test
Uncertainty
Inference

Keywords

  • bayesian network
  • bayes linear kinematics
  • engineering test strategy
  • reliability analysis
  • uncertainty quantification

Cite this

@article{f0c0c1620fb14390bc55b67d877a752b,
title = "Bayes linear Bayes graphical models in the design of optimal test strategies",
abstract = "Test and analysis plays a vital role in reducing uncertainty about the true performance of an engineering system. However tests can be expensive and designing an optimal test strategy can be challenging. We propose a Bayesian modelling process, which takes the form of a Bayesian Network, to determine anticipated test efficacy. Such a model supports engineering managers in assessing trade-offs between test resources and uncertainty reduction. Inference based on a full Bayesian model can be computationally demanding to the extent that it can limit practical application. To overcome this constraint, we develop a Bayes linear approximation for inference. This approach is known as a Bayes linear Bayes graphical model. After explaining the key principals of the method, we provide an application to a real industrial test to establish the condition of an ageing engineering system.",
keywords = "bayesian network, bayes linear kinematics, engineering test strategy, reliability analysis, uncertainty quantification",
author = "Kevin Wilson and John Quigley and Tim Bedford and Lesley Walls",
year = "2013",
month = "11",
language = "English",
volume = "9",
pages = "715--728",
journal = "International Journal of Performability Engineering",
issn = "0973-1318",
publisher = "RAMS Consultants",
number = "6",

}

TY - JOUR

T1 - Bayes linear Bayes graphical models in the design of optimal test strategies

AU - Wilson, Kevin

AU - Quigley, John

AU - Bedford, Tim

AU - Walls, Lesley

PY - 2013/11

Y1 - 2013/11

N2 - Test and analysis plays a vital role in reducing uncertainty about the true performance of an engineering system. However tests can be expensive and designing an optimal test strategy can be challenging. We propose a Bayesian modelling process, which takes the form of a Bayesian Network, to determine anticipated test efficacy. Such a model supports engineering managers in assessing trade-offs between test resources and uncertainty reduction. Inference based on a full Bayesian model can be computationally demanding to the extent that it can limit practical application. To overcome this constraint, we develop a Bayes linear approximation for inference. This approach is known as a Bayes linear Bayes graphical model. After explaining the key principals of the method, we provide an application to a real industrial test to establish the condition of an ageing engineering system.

AB - Test and analysis plays a vital role in reducing uncertainty about the true performance of an engineering system. However tests can be expensive and designing an optimal test strategy can be challenging. We propose a Bayesian modelling process, which takes the form of a Bayesian Network, to determine anticipated test efficacy. Such a model supports engineering managers in assessing trade-offs between test resources and uncertainty reduction. Inference based on a full Bayesian model can be computationally demanding to the extent that it can limit practical application. To overcome this constraint, we develop a Bayes linear approximation for inference. This approach is known as a Bayes linear Bayes graphical model. After explaining the key principals of the method, we provide an application to a real industrial test to establish the condition of an ageing engineering system.

KW - bayesian network

KW - bayes linear kinematics

KW - engineering test strategy

KW - reliability analysis

KW - uncertainty quantification

UR - http://www.ijpe-online.com/

M3 - Article

VL - 9

SP - 715

EP - 728

JO - International Journal of Performability Engineering

T2 - International Journal of Performability Engineering

JF - International Journal of Performability Engineering

SN - 0973-1318

IS - 6

ER -