A penalized local D-optimality approach to design for accelerated test models

Abstract The problem of choosing optimal levels of the acceleration variable for accelerated testing is an important issue in reliability analysis. Most recommendations have focused on minimizing the variance of an estimator of a particular characteristic, such as a percentile, for a specific parametric model. In this paper, a general approach based on “locally penalized” D-optimality (LPD-optimality) is proposed, which simultaneously minimizes the variances of the model parameter estimators. Application of the method is illustrated for inverse Gaussian-accelerated test models fitted to carbon fiber tensile strength data, where the fiber length is the “acceleration variable”.

[1]  W. J. Padgett A multiplicative damage model for strength of fibrous composite materials , 1998 .

[2]  S. D. Durham,et al.  Cumulative damage models for system failure with application to carbon fibers and composites , 1997 .

[3]  Arzu Onar,et al.  Inverse gaussian accelerated test models based on cumulative damage , 2000 .

[4]  K. Doksum,et al.  Models for variable-stress accelerated life testing experiments based on Wiener processes and the inverse Gaussian distribution , 1992 .

[5]  Luis A. Escobar,et al.  A Review of Recent Research and Current Issues in Accelerated Testing , 1993 .

[6]  N. Singpurwalla,et al.  Methods for Statistical Analysis of Reliability and Life Data. , 1975 .

[7]  Arzu Onar,et al.  Accelerated test models with the inverse Gaussian distribution , 2000 .

[8]  I. Ford,et al.  The Use of a Canonical Form in the Construction of Locally Optimal Designs for Non‐Linear Problems , 1992 .

[9]  I. Good Nonparametric roughness penalties for probability densities , 1971 .

[10]  D. Gaylor,et al.  Design for Optimal Prediction in Simple Linear Regression , 1965 .

[11]  Cuthbert Daniel,et al.  Design of Experiments for Most Precise Slope Estimation or Linear Extrapolation , 1950 .

[12]  J. Leroy Folks,et al.  The Inverse Gaussian Distribution , 1989 .

[13]  William Q. Meeker,et al.  Planning accelerated life tests with two or more experimental factors , 1995 .

[14]  G. K. Bhattacharyya,et al.  Inverse Gaussian regression and accelerated life tests , 1982 .

[15]  W. J. Padgett,et al.  Accelerated Test Models for System Strength Based on Birnbaum-Saunders Distributions , 1999, Lifetime data analysis.

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

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