Optimal release time on a stochastic single machine

Abstract

The job release times are usually treated as given information in most scheduling research. However, in practice, the release time can be a controlled variable. It should be controlled carefully for many reasons. Jobs released to the shop floor too early would result in carrying costs (e.g. cost incurred by having unnecessary capital tied up in inventory). On the other hand, jobs released too late would cause machines and other resources to be idle and miss the due dates. With proper job release control, the best trade-off can be achieved, In this paper we consider a single-machine scheduling problem in which the processing times are stochastic. In particular, two models are under consideration. Model I assumes no unreleased jobs in the system while Model II relaxes this assumption. Procedures are developed for both models to determine the optimal release time for the arriving job. The objective is to minimize the weighted average of the expected flow time and the expected make-span