Bayesian 1-Sample Prediction for the 2-Parameter Weibull Distribution

Abstract
Given the first few failures times of a sample of components whose lifetimes follow a Weibull distribution with both scale and shape parameters unknown, a Bayesian approach is used to derive prediction bounds for the failure times of the remaining components in the sample. Iterative procedures involving considerable computation are necessary for calculating prediction bounds, but some simplification is possible on restricting the shape parameter to a finite number of values. A numerical example illustrates the procedures.