Considering a warranty policy, an optimal truncated life test procedure is developed for the generalized exponential distribution. A total cost model, which involves the costs of testing, experimental time, lot acceptance or rejection, and the warranty, is adopted to establish the loss function of Bayesian decision method. A proper sample size of life test and an acceptance number of threshold are determined to minimize the expected total cost for lot decision. An example regarding the lifetime quality of integrated circuit products is used to demonstrate the proposed method. Numerical results have shown that the proposed method is beneficial for the reduction of expected total cost.
Journal of Industrial and Production Engineering 32(4), pp.225-231