Optimizing Stochastic Temporal Manufacturing Processes with Inventories: An Efficient Heuristic Algorithm Based on Deterministic Approximations

Download paper here

This paper deals with stochastic temporal manufacturing processes with work-in-process inventories in which multiple products are produced from raw materials and parts. The processes may be composed of subprocesses, which, in turn may be either composite or atomic, i.e., a machine on a manufacturing floor. We assume that machines' throughput is stochastic and so are work-in- process inventories and costs. We consider the problem of optimizing the process, that is, finding throughput expectation setting for each machine at each time point over the time horizon as to minimize the total cost of production subject to satisfying the production demand with a requested probability. To address this problem, we propose an efficient iterative heuristic algorithms that is based on (1) producing high quality candidate machine settings based on a deterministic approxima- tion of the stochastic problem, and (2) running stochastic simulations to find the best machine setting out of the produced candidates using optimal simulation budget allocation methods. We conduct an experimental study that shows that our algorithm significantly outperforms four popular simulation-based optimization algorithms."

Recommended citation: M. Krishnamoorthy, A. Brodsky, and D. Menasce, Optimizing Stochastic Temporal Manufacturing Processes with Inventories: An Efficient Heuristic Algorithm Based on Deterministic Approximations. In Proceedings of 2015 INFORMS Computing Soc. Conf., Richmond, VA. January 2015