Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability

Rafael Schwarzenegger; John Quigley; Lesley Walls

doi:10.3850/978-981-14-8593-0_4439-cd

Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability

Rafael Schwarzenegger, John Quigley, Lesley Walls

Management Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

39 Downloads (Pure)

Abstract

A method for estimating the reliability of a new product based on a comparative analysis of observed data for similar existing products has been motivated by an industry problem; see Quigley and Walls (to appear). Such estimates are required as part of contractual discussions with customers, as well as to inform the reliability program management for the new design. An empirical Bayes inference method has been developed based on a multivariate Poisson-Gamma probability model. The model aims to capture both aleatory and epistemic uncertainties. The latter are those which have the potential to be bought down by gathering more information such as learning about the existence of a potential design weaknesses. The model is multivariate since it involves modelling elements of the new product design reliability in relation to a set of relevant elements from multiple similar products. See, Quigley et al (2013) for more details of probability modelling of correlated events which provides a theoretical framework, and Quigley and Walls (2017) for an approach to construct prior distributions using empirical data.

Original language	English
Title of host publication	E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference
Subtitle of host publication	ESREL2020 PSAM15
Editors	Piero Baraldi, Francesco Di Maio, Enrico Zio
Place of Publication	Singapore
Pages	3022
Number of pages	1
ISBN (Electronic)	9789811485930
DOIs	https://doi.org/10.3850/978-981-14-8593-0_4439-cd
Publication status	Published - 5 Nov 2020
Event	The 30th European Safety and Reliability Conference and The 15th Probabilistic Safety Assessment and Management Conference - Online, Venice, Italy Duration: 1 Nov 2020 → 5 Nov 2020 https://www.esrel2020-psam15.org/

Conference

Conference	The 30th European Safety and Reliability Conference and The 15th Probabilistic Safety Assessment and Management Conference
Abbreviated title	ESREL 2020 PSAM 15
Country/Territory	Italy
City	Venice
Period	1/11/20 → 5/11/20
Internet address	https://www.esrel2020-psam15.org/

Keywords

empirical bayes inference
epistemic uncertainty
correlated event rates
new product design reliability
method evaluation

Access to Document

10.3850/978-981-14-8593-0_4439-cd

Schwarzenegger-etal-ESREL-2020-Evaluation-of-empirical-Bayes-estimators-of-correlated-event-rateAccepted author manuscript, 141 KB

Cite this

Schwarzenegger, R., Quigley, J., & Walls, L. (2020). Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability. In P. Baraldi, F. Di Maio, & E. Zio (Eds.), E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference: ESREL2020 PSAM15 (pp. 3022). https://doi.org/10.3850/978-981-14-8593-0_4439-cd

Schwarzenegger, Rafael ; Quigley, John ; Walls, Lesley. / Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability. E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference: ESREL2020 PSAM15. editor / Piero Baraldi ; Francesco Di Maio ; Enrico Zio. Singapore, 2020. pp. 3022

@inproceedings{d15fbae35e5f4721b3a39e11b2c3f3d1,

title = "Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability",

abstract = "A method for estimating the reliability of a new product based on a comparative analysis of observed data for similar existing products has been motivated by an industry problem; see Quigley and Walls (to appear). Such estimates are required as part of contractual discussions with customers, as well as to inform the reliability program management for the new design. An empirical Bayes inference method has been developed based on a multivariate Poisson-Gamma probability model. The model aims to capture both aleatory and epistemic uncertainties. The latter are those which have the potential to be bought down by gathering more information such as learning about the existence of a potential design weaknesses. The model is multivariate since it involves modelling elements of the new product design reliability in relation to a set of relevant elements from multiple similar products. See, Quigley et al (2013) for more details of probability modelling of correlated events which provides a theoretical framework, and Quigley and Walls (2017) for an approach to construct prior distributions using empirical data.",

keywords = "empirical bayes inference, epistemic uncertainty, correlated event rates, new product design reliability, method evaluation",

author = "Rafael Schwarzenegger and John Quigley and Lesley Walls",

year = "2020",

month = nov,

day = "5",

doi = "10.3850/978-981-14-8593-0_4439-cd",

language = "English",

pages = "3022",

editor = "Piero Baraldi and {Di Maio}, Francesco and Enrico Zio",

booktitle = "E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference",

note = "The 30th European Safety and Reliability Conference and The 15th Probabilistic Safety Assessment and Management Conference, ESREL 2020 PSAM 15 ; Conference date: 01-11-2020 Through 05-11-2020",

url = "https://www.esrel2020-psam15.org/",

}

Schwarzenegger, R, Quigley, J & Walls, L 2020, Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability. in P Baraldi, F Di Maio & E Zio (eds), E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference: ESREL2020 PSAM15. Singapore, pp. 3022, The 30th European Safety and Reliability Conference and The 15th Probabilistic Safety Assessment and Management Conference, Venice, Italy, 1/11/20. https://doi.org/10.3850/978-981-14-8593-0_4439-cd

Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability. / Schwarzenegger, Rafael; Quigley, John ; Walls, Lesley.
E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference: ESREL2020 PSAM15. ed. / Piero Baraldi; Francesco Di Maio; Enrico Zio. Singapore, 2020. p. 3022.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability

AU - Schwarzenegger, Rafael

AU - Quigley, John

AU - Walls, Lesley

PY - 2020/11/5

Y1 - 2020/11/5

N2 - A method for estimating the reliability of a new product based on a comparative analysis of observed data for similar existing products has been motivated by an industry problem; see Quigley and Walls (to appear). Such estimates are required as part of contractual discussions with customers, as well as to inform the reliability program management for the new design. An empirical Bayes inference method has been developed based on a multivariate Poisson-Gamma probability model. The model aims to capture both aleatory and epistemic uncertainties. The latter are those which have the potential to be bought down by gathering more information such as learning about the existence of a potential design weaknesses. The model is multivariate since it involves modelling elements of the new product design reliability in relation to a set of relevant elements from multiple similar products. See, Quigley et al (2013) for more details of probability modelling of correlated events which provides a theoretical framework, and Quigley and Walls (2017) for an approach to construct prior distributions using empirical data.

AB - A method for estimating the reliability of a new product based on a comparative analysis of observed data for similar existing products has been motivated by an industry problem; see Quigley and Walls (to appear). Such estimates are required as part of contractual discussions with customers, as well as to inform the reliability program management for the new design. An empirical Bayes inference method has been developed based on a multivariate Poisson-Gamma probability model. The model aims to capture both aleatory and epistemic uncertainties. The latter are those which have the potential to be bought down by gathering more information such as learning about the existence of a potential design weaknesses. The model is multivariate since it involves modelling elements of the new product design reliability in relation to a set of relevant elements from multiple similar products. See, Quigley et al (2013) for more details of probability modelling of correlated events which provides a theoretical framework, and Quigley and Walls (2017) for an approach to construct prior distributions using empirical data.

KW - empirical bayes inference

KW - epistemic uncertainty

KW - correlated event rates

KW - new product design reliability

KW - method evaluation

UR - https://www.esrel2020-psam15.org/index.html

U2 - 10.3850/978-981-14-8593-0_4439-cd

DO - 10.3850/978-981-14-8593-0_4439-cd

M3 - Conference contribution book

SP - 3022

BT - E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference

A2 - Baraldi, Piero

A2 - Di Maio, Francesco

A2 - Zio, Enrico

CY - Singapore

T2 - The 30th European Safety and Reliability Conference and The 15th Probabilistic Safety Assessment and Management Conference

Y2 - 1 November 2020 through 5 November 2020

ER -

Schwarzenegger R, Quigley J , Walls L. Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability. In Baraldi P, Di Maio F, Zio E, editors, E-proceedings of the 30th European Safety and Reliability Conference and 15th Probabilistic Safety Assessment and Management Conference: ESREL2020 PSAM15. Singapore. 2020. p. 3022 doi: 10.3850/978-981-14-8593-0_4439-cd

Evaluation of empirical Bayes estimators of correlated event rates and implications for predicting new product reliability

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

A methodology for constructing subjective probability distributions with data

A Bayes linear Bayes method for estimation of correlated event rates

Cite this