Iwein @ mit.edu A Simple and Effective Component Procurement Policy for Stochastic Assembly Systems

This paper examines the problem of optimally procuring components in a single-item stochastic assem-bly system that operates in a make-to-stock fashion. The suppliers are uncapacitated and the components have independent but non-identically distributed stochastic procurement lead times. Assembly is instan-taneous, product demand follows a Poisson process and unsatisfied demand is backordered. The following class of policies is considered: the finished goods inventory is initially filled to its base stock level, and each customer order triggers a replenishment order for all components after a component-dependent postpone-ment lead time. The objective is to minimize the sum of steady-state holding and backorder costs over this class of replenishment policies. To keep the analysis tractable, we impose a synchronization assumption that no mixing occurs between sets of component orders. Combining existing results from queueing theory with original results concerning probability distributions that are closed under maximization and translation, we derive a simple approximate solution to the problem when lead time variances are identical. In a simulation study that uses data from a Hewlett-Packard facility, our derived policy is within 2 % of optimal and signifi-cantly outperforms policies that ignore either component dependence or the stochasticity in the lead times. In addition, we show that the policy is quite robust with respect to various model assumptions, except the synchronization one. 1