Data analysis for accelerated life tests with constrained randomization

Kangwon Seo, Rong Pan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Accelerated life tests (ALTs) often involve experimental protocols with constrained randomization such as subsampling or random block. As a result, life-time data may construct a grouped structure among the observations. In this paper, we develop a generalized linear mixed model (GLMM) approach for analyzing ALT data with a grouped structure in order to reflect random effects of groups in the model. The GLMM approach provides a flexible way to model censored failure time data with random effects. Particularly, for Weibull failure time distribution, we describe an iterative procedure for the model parameters estimation and derive the asymptotic variance-covariance matrix using the approximated likelihood function. Two examples of life-time data with subsampling and random block are analyzed by the proposed method, which is implemented by modern computer software.

Original languageEnglish (US)
Title of host publicationProceedings - Annual Reliability and Maintainability Symposium
PublisherInstitute of Electrical and Electronics Engineers Inc.
Volume2016-April
ISBN (Print)9781509002481
DOIs
StatePublished - Apr 5 2016
EventAnnual Reliability and Maintainability Symposium, RAMS 2016 - Tucson, United States
Duration: Jan 25 2016Jan 28 2016

Other

OtherAnnual Reliability and Maintainability Symposium, RAMS 2016
CountryUnited States
CityTucson
Period1/25/161/28/16

Fingerprint

Accelerated Life Test
Randomisation
Data analysis
Generalized Linear Mixed Model
Subsampling
Random Effects
Lifetime
Failure Time Data
Variance-covariance Matrix
Failure Time
Asymptotic Variance
Weibull
Censored Data
Iterative Procedure
Likelihood Function
Parameter Estimation
Model
Covariance matrix
Parameter estimation
Network protocols

Keywords

  • Accelerated life test
  • generalized linear mixed model
  • maximum likelihood estimation
  • random effects

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Mathematics(all)
  • Computer Science Applications

Cite this

Seo, K., & Pan, R. (2016). Data analysis for accelerated life tests with constrained randomization. In Proceedings - Annual Reliability and Maintainability Symposium (Vol. 2016-April). [7447962] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RAMS.2016.7447962

Data analysis for accelerated life tests with constrained randomization. / Seo, Kangwon; Pan, Rong.

Proceedings - Annual Reliability and Maintainability Symposium. Vol. 2016-April Institute of Electrical and Electronics Engineers Inc., 2016. 7447962.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Seo, K & Pan, R 2016, Data analysis for accelerated life tests with constrained randomization. in Proceedings - Annual Reliability and Maintainability Symposium. vol. 2016-April, 7447962, Institute of Electrical and Electronics Engineers Inc., Annual Reliability and Maintainability Symposium, RAMS 2016, Tucson, United States, 1/25/16. https://doi.org/10.1109/RAMS.2016.7447962
Seo K, Pan R. Data analysis for accelerated life tests with constrained randomization. In Proceedings - Annual Reliability and Maintainability Symposium. Vol. 2016-April. Institute of Electrical and Electronics Engineers Inc. 2016. 7447962 https://doi.org/10.1109/RAMS.2016.7447962
Seo, Kangwon ; Pan, Rong. / Data analysis for accelerated life tests with constrained randomization. Proceedings - Annual Reliability and Maintainability Symposium. Vol. 2016-April Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{d64c2c781bd549e081f3bb6ad67d4467,
title = "Data analysis for accelerated life tests with constrained randomization",
abstract = "Accelerated life tests (ALTs) often involve experimental protocols with constrained randomization such as subsampling or random block. As a result, life-time data may construct a grouped structure among the observations. In this paper, we develop a generalized linear mixed model (GLMM) approach for analyzing ALT data with a grouped structure in order to reflect random effects of groups in the model. The GLMM approach provides a flexible way to model censored failure time data with random effects. Particularly, for Weibull failure time distribution, we describe an iterative procedure for the model parameters estimation and derive the asymptotic variance-covariance matrix using the approximated likelihood function. Two examples of life-time data with subsampling and random block are analyzed by the proposed method, which is implemented by modern computer software.",
keywords = "Accelerated life test, generalized linear mixed model, maximum likelihood estimation, random effects",
author = "Kangwon Seo and Rong Pan",
year = "2016",
month = "4",
day = "5",
doi = "10.1109/RAMS.2016.7447962",
language = "English (US)",
isbn = "9781509002481",
volume = "2016-April",
booktitle = "Proceedings - Annual Reliability and Maintainability Symposium",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Data analysis for accelerated life tests with constrained randomization

AU - Seo, Kangwon

AU - Pan, Rong

PY - 2016/4/5

Y1 - 2016/4/5

N2 - Accelerated life tests (ALTs) often involve experimental protocols with constrained randomization such as subsampling or random block. As a result, life-time data may construct a grouped structure among the observations. In this paper, we develop a generalized linear mixed model (GLMM) approach for analyzing ALT data with a grouped structure in order to reflect random effects of groups in the model. The GLMM approach provides a flexible way to model censored failure time data with random effects. Particularly, for Weibull failure time distribution, we describe an iterative procedure for the model parameters estimation and derive the asymptotic variance-covariance matrix using the approximated likelihood function. Two examples of life-time data with subsampling and random block are analyzed by the proposed method, which is implemented by modern computer software.

AB - Accelerated life tests (ALTs) often involve experimental protocols with constrained randomization such as subsampling or random block. As a result, life-time data may construct a grouped structure among the observations. In this paper, we develop a generalized linear mixed model (GLMM) approach for analyzing ALT data with a grouped structure in order to reflect random effects of groups in the model. The GLMM approach provides a flexible way to model censored failure time data with random effects. Particularly, for Weibull failure time distribution, we describe an iterative procedure for the model parameters estimation and derive the asymptotic variance-covariance matrix using the approximated likelihood function. Two examples of life-time data with subsampling and random block are analyzed by the proposed method, which is implemented by modern computer software.

KW - Accelerated life test

KW - generalized linear mixed model

KW - maximum likelihood estimation

KW - random effects

UR - http://www.scopus.com/inward/record.url?scp=84968784516&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84968784516&partnerID=8YFLogxK

U2 - 10.1109/RAMS.2016.7447962

DO - 10.1109/RAMS.2016.7447962

M3 - Conference contribution

AN - SCOPUS:84968784516

SN - 9781509002481

VL - 2016-April

BT - Proceedings - Annual Reliability and Maintainability Symposium

PB - Institute of Electrical and Electronics Engineers Inc.

ER -