Abstract
This paper presents an economic life test acceptance sampling plan using item-censored data in a Bayesian situation. It is assumed that failures in a life test are replaced immediately by new ones. A prior distribution is assigned to the mean lifetime θ for the calculation of the expected total cost. Then the optimum plan is chosen to be the one which minimizes the expected total cost. A direct search method and a dual programming method are introduced, with emphasis on the latter. A numerical example is presented to illustrate the procedure. A sensitivity study is included on the effect of a wrong choice of the prior distribution.