![IconGems-Print](images/IconGems-Print.png)
333 – Topics in Reliability, Data Visualization and Modeling
Bayesian Prediction Bounds in Accelerated Life Testing: Weibull Models with Two Levels of Acceleration
Ananda Jayawardhana
Pittsburg State University
Bayesian lower prediction bounds for a future observation from a Weibull distribution at the design level of stress using Type II censored data from two levels of accelerated stress is considered. The scale parameter of the Weibull distribution is assumed to have an inverse power relationship with the levels of stress while the shape parameter is assumed to be a constant. OpenBUGS is used to calculate Bayesian estimates of the parameters in a simulation study. A previous simulation study (Jayawardhana and Samaranayake, 2003) provides comparable results using Maximum Likelihood Predictive Density Method. Results of the current simulation study is compared with the results of the study by Jayawardhana and Samaranayake (2003). Proposed method will be illustrated through a well-known data set on breakdown times for insulating fluids (Nelson, 1972).