Many quality control problems involve the estimation of proportion defective in a lot. The classical methods assume no prior knowledge of the upper bound on proportion defective. This paper addresses the problem of the determination of confidence intervals for proportion defective for cases where an upper bound on the proportion defective can be assumed. The tables presented, developed using a uniform prior on the proportion defective, reflect assumptions of both large lot size, using the binomial distribution, and small lot sizes (N= 25,N= 50,N= 100), using the hypergeometric distribution. These tables give upper Bayesian confidence limits on the proportion defective.