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

The Weibull log-linear model is a widely-used accelerated life-test model i n reliability engineering, The standard deviation of log(life), s, was ofte n assumed to be a constant or stress-independent. However, theoretical & ex perimental research results suggest that, in many cases, s is stress-depend ent. The data analysis via the MLE method must be performed numerically, be cause of the complexity of the model and many unknown parameters being invo lved, The commonly-used methods often fail to converge when the starting po int 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 f acilitate this process, an efficient algorithm is presented in this paper, to obtain the MLE of the model parameters from test data (with or without c ensoring) for both stress-independent and stress-dependent models, The vali dity & effectiveness of this procedure are illustrated with numerical examp les. The method is numerically stable, and easy to implement & program.