Inventory policies in dual sourcing systems with uncertain yield

Abstract

We study a spare part supply system where both Traditional Manufacturing and Additive Manufacturing (also know as 3D printing) are used for replenishment of the spare parts inventories. Demands for the spare parts occur according to a Poisson process, and the failed parts are immediately replaced from the inventory, if available. If inventory is not available, items are backordered and fulfilled when a spare becomes available (i.e., a replenishment is received from one of the suppliers). Additive Manufacturing offers the advantage of shorter lead times, however, at higher production costs. Moreover, Additive Manufacturing processes often have uncertain yield, leading to the fact that not every produced part will satisfy the quality control and can be used to replenish the inventory. In this paper, we propose a Linear Programming (LP) based optimization problem that decides which of the processes to use for replenishment of the inventory while minimizing the total (holding + backorder) system costs.