Abstract

Step-Stress Accelerated Life Testing (SSALT) is a special type of experiment that tests a product’s lifetime with time-varying stress levels. Typical testing protocols deployed in SSALTs cannot implement complete randomization of experiments; instead, they often result in grouped structures of experimental units and, thus, correlated observations. In this article, we propose a Generalized Linear Mixed Model (GLMM) approach to take into account the random group effect in SSALT. Failure times are assumed to be exponentially distributed under any stress level. Two parameter estimation methods, Adaptive Gaussian Quadrature (AGQ) and Integrated Nested Laplace Approximation (INLA), are introduced. A simulation study is conducted to compare the proposed random effect modelwith the traditional model, which pools data groups together, and with the fixed effect model.We also compareAGQand INLA with different priors for parameter estimation. Results show that the proposed model can validate the existence of group-to-group variation. Lastly, the GLMM model is applied to a real data and it shows that disregarding experimental protocols in SSALT may result in large bias in the estimation of the effect of stress variable.

Original languageEnglish (US)
Pages (from-to)885-898
Number of pages14
JournalIISE Transactions
Volume49
Issue number9
DOIs
StatePublished - 2017

Fingerprint

Testing
Parameter estimation
Experiments

Keywords

  • Adaptive Gaussian quadrature
  • Constrained randomization
  • Generalized linear mixed model
  • Integrated nested Laplace approximation
  • Step-stress accelerated life test

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Data analysis of step-stress accelerated life tests with heterogeneous group effects. / Seo, Kangwon; Pan, Rong.

In: IISE Transactions, Vol. 49, No. 9, 2017, p. 885-898.

Research output: Contribution to journalArticle

@article{bb11e3d377fb40f88a9965569a5fe311,
title = "Data analysis of step-stress accelerated life tests with heterogeneous group effects",
abstract = "Step-Stress Accelerated Life Testing (SSALT) is a special type of experiment that tests a product’s lifetime with time-varying stress levels. Typical testing protocols deployed in SSALTs cannot implement complete randomization of experiments; instead, they often result in grouped structures of experimental units and, thus, correlated observations. In this article, we propose a Generalized Linear Mixed Model (GLMM) approach to take into account the random group effect in SSALT. Failure times are assumed to be exponentially distributed under any stress level. Two parameter estimation methods, Adaptive Gaussian Quadrature (AGQ) and Integrated Nested Laplace Approximation (INLA), are introduced. A simulation study is conducted to compare the proposed random effect modelwith the traditional model, which pools data groups together, and with the fixed effect model.We also compareAGQand INLA with different priors for parameter estimation. Results show that the proposed model can validate the existence of group-to-group variation. Lastly, the GLMM model is applied to a real data and it shows that disregarding experimental protocols in SSALT may result in large bias in the estimation of the effect of stress variable.",
keywords = "Adaptive Gaussian quadrature, Constrained randomization, Generalized linear mixed model, Integrated nested Laplace approximation, Step-stress accelerated life test",
author = "Kangwon Seo and Rong Pan",
year = "2017",
doi = "10.1080/24725854.2017.1312038",
language = "English (US)",
volume = "49",
pages = "885--898",
journal = "IISE Transactions",
issn = "2472-5854",
publisher = "Taylor and Francis Ltd.",
number = "9",

}

TY - JOUR

T1 - Data analysis of step-stress accelerated life tests with heterogeneous group effects

AU - Seo, Kangwon

AU - Pan, Rong

PY - 2017

Y1 - 2017

N2 - Step-Stress Accelerated Life Testing (SSALT) is a special type of experiment that tests a product’s lifetime with time-varying stress levels. Typical testing protocols deployed in SSALTs cannot implement complete randomization of experiments; instead, they often result in grouped structures of experimental units and, thus, correlated observations. In this article, we propose a Generalized Linear Mixed Model (GLMM) approach to take into account the random group effect in SSALT. Failure times are assumed to be exponentially distributed under any stress level. Two parameter estimation methods, Adaptive Gaussian Quadrature (AGQ) and Integrated Nested Laplace Approximation (INLA), are introduced. A simulation study is conducted to compare the proposed random effect modelwith the traditional model, which pools data groups together, and with the fixed effect model.We also compareAGQand INLA with different priors for parameter estimation. Results show that the proposed model can validate the existence of group-to-group variation. Lastly, the GLMM model is applied to a real data and it shows that disregarding experimental protocols in SSALT may result in large bias in the estimation of the effect of stress variable.

AB - Step-Stress Accelerated Life Testing (SSALT) is a special type of experiment that tests a product’s lifetime with time-varying stress levels. Typical testing protocols deployed in SSALTs cannot implement complete randomization of experiments; instead, they often result in grouped structures of experimental units and, thus, correlated observations. In this article, we propose a Generalized Linear Mixed Model (GLMM) approach to take into account the random group effect in SSALT. Failure times are assumed to be exponentially distributed under any stress level. Two parameter estimation methods, Adaptive Gaussian Quadrature (AGQ) and Integrated Nested Laplace Approximation (INLA), are introduced. A simulation study is conducted to compare the proposed random effect modelwith the traditional model, which pools data groups together, and with the fixed effect model.We also compareAGQand INLA with different priors for parameter estimation. Results show that the proposed model can validate the existence of group-to-group variation. Lastly, the GLMM model is applied to a real data and it shows that disregarding experimental protocols in SSALT may result in large bias in the estimation of the effect of stress variable.

KW - Adaptive Gaussian quadrature

KW - Constrained randomization

KW - Generalized linear mixed model

KW - Integrated nested Laplace approximation

KW - Step-stress accelerated life test

UR - http://www.scopus.com/inward/record.url?scp=85028654763&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028654763&partnerID=8YFLogxK

U2 - 10.1080/24725854.2017.1312038

DO - 10.1080/24725854.2017.1312038

M3 - Article

AN - SCOPUS:85028654763

VL - 49

SP - 885

EP - 898

JO - IISE Transactions

JF - IISE Transactions

SN - 2472-5854

IS - 9

ER -