Accelerated Life Test Plans Robust to Misspecification of the Stress—Life Relationship

This article discusses methodology for deriving the asymptotic distribution of maximum likelihood estimators of model parameters in accelerated life tests (ALTs) when the form of the stress–life relationship is misspecified. If the practitioner can describe possible departures from an assumed ALT model, often linear in the appropriately transformed accelerating variable, then the results here can be used to obtain ALT plans that provide protection against potential bias and loss in estimation efficiency. General results are derived for log location-scale distributions and time-censored data, whereas specific results are presented for the lognormal and Weibull distributions. In a practical ALT situation, the methodology is applied to derive robust test plans based on minimizing asymptotic bias, asymptotic variance, or asymptotic mean squared error. Test plans are evaluated in terms of bias, efficiency, extrapolation to use level, and expected number failing.

[1]  Herman Chernoff,et al.  Optimal Accelerated Life Designs for Estimation , 1962 .

[2]  G. Box,et al.  A Basis for the Selection of a Response Surface Design , 1959 .

[3]  Francis G. Pascual,et al.  Model-Robust Test Plans With Applications in Accelerated Life Testing , 2003, Technometrics.

[4]  John E. Dennis,et al.  Algorithm 573: NL2SOL—An Adaptive Nonlinear Least-Squares Algorithm [E4] , 1981, TOMS.

[5]  W. Nelson.  Statistical Methods for Reliability Data , 1998 .

[6]  Paula Kanarek,et al.  Volume 10: How to Plan an Accelerated Life Test—Some Practical Guidelines , 1987 .

[7]  Holger Dette,et al.  On a mixture of the D- and D1-optimality criterion in polynomial regression , 1993 .

[8]  H. White.  Maximum Likelihood Estimation of Misspecified Models , 1982 .

[9]  Francis G. Pascual,et al.  Lognormal and Weibull accelerated life test plans under distribution misspecification , 2005, IEEE Transactions on Reliability.

[10]  H. Dette,et al.  A Generalization of $D$- and $D_1$-Optimal Designs in Polynomial Regression , 1990 .

[11]  Marvin Zelen.  [Linear Estimation of a Regression Relationship from Censored Data Part I: Simple Methods and Their Application]: Discussion No. 1 , 1972 .

[12]  N. L. Johnson,et al.  Continuous Univariate Distributions. , 1995 .

[13]  Thomas J. Kielpinski,et al.  Optimum Censored Accelerated Life Tests for Normal and Lognormal Life Distributions , 1975, IEEE Transactions on Reliability.

[14]  William I. Notz,et al.  OPTIMAL DESIGNS FOR REGRESSION MODELS WITH POSSIBLE BIAS , 1989 .

[15]  Josep Ginebra,et al.  Minimax approach to accelerated life tests , 1998 .

[16]  W. B. Nelson,et al.  A bibliography of accelerated test plans , 2005, IEEE Transactions on Reliability.

[17]  H. Chernoff.  Locally Optimal Designs for Estimating Parameters , 1953 .

[18]  Michael J. Phillips,et al.  Statistical Methods for Reliability Data Analysis , 2003 .

[19]  W. Meeker.  A Comparison of Accelerated Life Test Plans for Weibull and Lognormal Distributions and Type I Censoring , 1984 .

[20]  Friedrich Pukelsheim,et al.  E-Optimal Designs for Polynomial Regression , 1993 .

[21]  Douglas P. Wiens,et al.  Minimax regression designs for approximately linear models with autocorrelated errors , 1996 .

[22]  Stephen M. Stigler,et al.  Optimal Experimental Design for Polynomial Regression , 1971 .

[23]  K. Chaloner,et al.  Bayesian design for accelerated life testing , 1992 .

[24]  Holger Dette,et al.  Robust designs for multivariate polynomial regression on the d-cube , 1994 .

[25]  E. Läuter.  Experimental design in a class of models , 1974 .

[26]  W. Meeker.  Accelerated Testing: Statistical Models, Test Plans, and Data Analyses , 1991 .

[27]  Francis Pascual,et al.  On minimax designs when there are two candidate models , 2002 .

[28]  William Q. Meeker,et al.  Use of Sensitivity Analysis to Assess the Effect of Model Uncertainty in Analyzing Accelerated Life Test Data , 2003 .

[29]  William Q. Meeker,et al.  Theory for Optimum Accelerated Censored Life Tests for Weibull and Extreme Value Distributions , 1978 .

[30]  Gerald J. Hahn,et al.  Linear Estimation of a Regression Relationship from Censored Data—Part II Best Linear Unbiased Estimation and Theory , 1973 .

[31]  George E. P. Box,et al.  Empirical Model‐Building and Response Surfaces , 1988 .

[32]  Grace Montepiedra,et al.  Minimum bias designs with constraints , 1997 .

[33]  Douglas P. Wiens,et al.  Robust designs for approximately polynomial regression , 1997 .