Control policies for multi-product multi-stage manufacturing systems: An experimental approach

We investigate the behaviour of a multi-product, multi-stage pull-type production/inventory system operating under a variety of manufacturing control policies. The system consists of multiple stages in series with finite interstage buffer space and a single processor. The first stage has always sufficient amount of raw material to work on and the last stage is followed by a finite-capacity warehouse. Demand for a product follows a compound Poisson distribution. Any unsatisfied portion of the demand is backordered. The inventory level as well as the production process of each product at each stage are controlled through the use of an (R, r) continuous-review inventory control policy. A set-up is involved whenever a switch-over takes place to another product at any stage. Using a particular test case, we have evaluated various manufacturing strategies and switching policies under static as well as dynamic priority structures. We have compared the operating regimes by observing a set of performance measures such as the average inventory levels, the average backorder size, the average backorder levels, the probability of backorder and the average number of set-ups. We conclude proposing control policies for better customer service level for given scenarios.