An improved ordering policy for continuous review inventory systems with arbitrary inter-demand time distributions

In this paper, we revisit the optimality of (s,S) policies in continuous re view inventory models where demand forms a renewal process. We explain why when orders are placed at demand epochs, (s,S) policies are not optimal in general, and propose a simple but practical ordering policy by introducing a delay in order placement as a policy parameter. Under our proposed policy , the operating characteristics of such systems can be evaluated using the existing results in the literature. In order to demonstrate the effectivene ss of our policy, we restrict our analysis to a special class of inventory systems where demand follows a counting process, fixed ordering costs are n egligible and leadtimes are constant. We derive expressions for the operati ng characteristics of such systems under our policy and present insights on the behavior of the policy parameters which minimize the average total cos t rate under our policy. Using these results, we develop an efficient heuri stic for computing the policy parameters. Finally, we investigate the effec tiveness of our policy in comparison to policies which place orders only at demand epochs. The results of our numerical experiment indicate that our p olicy can result in a significant savings.