Fitting the Weibull log-linear model to accelerated life-test data

The Weibull log-linear model is a widely-used accelerated life-test model in reliability engineering. The standard deviation of log(life), s, was often assumed to be a constant or stress-independent. However, theoretical and experimental research results suggest that, in many cases, s is stress-dependent. The data analysis via the MLE method must be performed numerically, because of the complexity of the model and many unknown parameters being involved. The commonly-used methods often fail to converge when the starting point is not close to the solution, especially for censored data. Generally, no easy-to-use software is available for the Weibull log-linear model. To facilitate this process, an efficient algorithm is presented in this paper, to obtain the MLE of the model parameters from test data (with or without censoring) for both stress-independent and stress-dependent models. The validity and effectiveness of this procedure are illustrated with numerical examples. The method is numerically stable, and easy to implement and program.

[1]  J. Engel,et al.  A generalized linear modeling approach to robust design , 1996 .

[2]  P. McCullagh,et al.  Generalized Linear Models , 1972, Predictive Analytics.

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

[4]  Sheng-Tsaing Tseng,et al.  A classification rule arising from accelerated life test , 1994 .

[5]  Jerald F. Lawless,et al.  Statistical Models and Methods for Lifetime Data. , 1983 .

[6]  J. Lawless.  Statistical Models and Methods for Lifetime Data , 2002 .

[7]  Hideo Hirose,et al.  Estimation of threshold stress in accelerated life-testing , 1993 .

[8]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[9]  C. Chan,et al.  Temperature-dependent standard deviation of log(failure time) distributions , 1991 .

[10]  Enrico A. Colosimo,et al.  Accelerated life tests analyzed by a piecewise exponential distribution via generalized linear models , 1996, IEEE Trans. Reliab..

[11]  Alan Watkins,et al.  Review: likelihood method for fitting Weibull log-linear models to accelerated life-test data , 1994 .

[12]  Martin Newby.  Perspective on Weibull proportional-hazards models , 1994 .

[13]  J. A. Schwarz.  Effect of temperature on the variance of the log‐normal distribution of failure times due to electromigration damage , 1987 .

[14]  R. L. Hartman,et al.  Methodology of accelerated aging , 1985, AT&T Technical Journal.

[15]  S. Lauritzen,et al.  Globally convergent algorithms for maximizing a likelihood function , 1991 .

[16]  W. J. Kolarik,et al.  On/off cycling under multiple stresses , 1989 .

[17]  Francisco Louzada-Neto,et al.  Analysis of accelerated life tests with weibull failure distribution via generalized linear models , 1994 .

[18]  William Q. Meeker,et al.  Optimum Accelerated Life Tests Wth a Nonconstant Scale Parameter , 1994 .

[19]  J. A. McLinn.  New analysis methods of multilevel accelerated life tests , 1999, Annual Reliability and Maintainability. Symposium. 1999 Proceedings (Cat. No.99CH36283).

[20]  David M. Miller,et al.  Improvement of an industrial thermostat using designed experiments , 1993 .

[21]  Helmut Schneider.  Reliability & Life Testing Handbook , 1994 .

[22]  P. Green.  Iteratively reweighted least squares for maximum likelihood estimation , 1984 .

[23]  W Nelson,et al.  Fitting of fatigue curves with nonconstant standard deviation to data with runouts , 1984 .