Bayes linear graphical models in the design of optimal test strategies

Research output: Contribution to conferencePaper

Abstract

Testing plays a vital role in reducing uncertainty but is resource intensive and identifying the best design is a difficult process. During the development of a system there are a number of potential tests that can be performed with varying efficacy and resource requirements. In this paper we propose a Bayesian modelling process which takes the form of a Bayesian Belief Network (BBN) to determine test efficacy and permits programme managers to assess optimal trade-offs between uncertainty reduction and resources. Supporting inference from a full Bayesian model can be prohibitively expensive computationally so we utilise a Bayes linear approximation, known as a Bayes linear Bayes graphical model, to the inference.

Conference

Conference8th International Conference on Mathematical Methods in Reliability (MMR2013)
CountrySouth Africa
CityStellenboch
Period1/07/134/07/13

Fingerprint

Graphical models
Resources
Optimal test
Inference
Efficacy
Uncertainty
Managers
Testing
Bayesian modeling
Bayesian belief networks
Bayesian model
Trade-offs
Approximation

Keywords

  • Bayesian modelling
  • Bayes graphical model
  • Bayesian belief networks

Cite this

Wilson, K., Quigley, J., Bedford, T., & Walls, L. (2013). Bayes linear graphical models in the design of optimal test strategies. Paper presented at 8th International Conference on Mathematical Methods in Reliability (MMR2013), Stellenboch, South Africa.
Wilson, Kevin ; Quigley, John ; Bedford, Tim ; Walls, Lesley. / Bayes linear graphical models in the design of optimal test strategies. Paper presented at 8th International Conference on Mathematical Methods in Reliability (MMR2013), Stellenboch, South Africa.
@conference{fbacc5c7dfcb4ef1bc0f7c5b2a845cb2,
title = "Bayes linear graphical models in the design of optimal test strategies",
abstract = "Testing plays a vital role in reducing uncertainty but is resource intensive and identifying the best design is a difficult process. During the development of a system there are a number of potential tests that can be performed with varying efficacy and resource requirements. In this paper we propose a Bayesian modelling process which takes the form of a Bayesian Belief Network (BBN) to determine test efficacy and permits programme managers to assess optimal trade-offs between uncertainty reduction and resources. Supporting inference from a full Bayesian model can be prohibitively expensive computationally so we utilise a Bayes linear approximation, known as a Bayes linear Bayes graphical model, to the inference.",
keywords = "Bayesian modelling, Bayes graphical model, Bayesian belief networks",
author = "Kevin Wilson and John Quigley and Tim Bedford and Lesley Walls",
year = "2013",
month = "7",
language = "English",
note = "8th International Conference on Mathematical Methods in Reliability (MMR2013) ; Conference date: 01-07-2013 Through 04-07-2013",

}

Wilson, K, Quigley, J, Bedford, T & Walls, L 2013, 'Bayes linear graphical models in the design of optimal test strategies' Paper presented at 8th International Conference on Mathematical Methods in Reliability (MMR2013), Stellenboch, South Africa, 1/07/13 - 4/07/13, .

Bayes linear graphical models in the design of optimal test strategies. / Wilson, Kevin; Quigley, John; Bedford, Tim; Walls, Lesley.

2013. Paper presented at 8th International Conference on Mathematical Methods in Reliability (MMR2013), Stellenboch, South Africa.

Research output: Contribution to conferencePaper

TY - CONF

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

AU - Wilson, Kevin

AU - Quigley, John

AU - Bedford, Tim

AU - Walls, Lesley

PY - 2013/7

Y1 - 2013/7

N2 - Testing plays a vital role in reducing uncertainty but is resource intensive and identifying the best design is a difficult process. During the development of a system there are a number of potential tests that can be performed with varying efficacy and resource requirements. In this paper we propose a Bayesian modelling process which takes the form of a Bayesian Belief Network (BBN) to determine test efficacy and permits programme managers to assess optimal trade-offs between uncertainty reduction and resources. Supporting inference from a full Bayesian model can be prohibitively expensive computationally so we utilise a Bayes linear approximation, known as a Bayes linear Bayes graphical model, to the inference.

AB - Testing plays a vital role in reducing uncertainty but is resource intensive and identifying the best design is a difficult process. During the development of a system there are a number of potential tests that can be performed with varying efficacy and resource requirements. In this paper we propose a Bayesian modelling process which takes the form of a Bayesian Belief Network (BBN) to determine test efficacy and permits programme managers to assess optimal trade-offs between uncertainty reduction and resources. Supporting inference from a full Bayesian model can be prohibitively expensive computationally so we utilise a Bayes linear approximation, known as a Bayes linear Bayes graphical model, to the inference.

KW - Bayesian modelling

KW - Bayes graphical model

KW - Bayesian belief networks

UR - http://www.sastat.org.za/mmr2013/Programme-MMR-PAGES11-14.pdf

M3 - Paper

ER -

Wilson K, Quigley J, Bedford T, Walls L. Bayes linear graphical models in the design of optimal test strategies. 2013. Paper presented at 8th International Conference on Mathematical Methods in Reliability (MMR2013), Stellenboch, South Africa.