Modified Base Stock Policies for Capacitated Systems with Multiple Suppliers

In this research, effective inventory models are developed to manage single-product supply chain systems in which there are two capacitated suppliers with certain attributes. These models consider an infinite time horizon with independent and identically distributed demand in each period. The characteristics of interest in developing the models are setup cost, unit price, and capacity.This thesis develops modified base stock models for two cases regarding the characteristics of each supplier. In Case 1, the suppliers are distinguished by one supplier having a lower setup cost, a higher unit price, and a higher capacity. In Case 2, the suppliers are distinguished by one supplier having a lower setup cost, a lower unit price, and a lower capacity. For each policy, we develop a set of three order point levels to guide the manufacturer on how to order. The manufacturer can either order from the first supplier, the second supplier, both, or neither. For each case, we develop a Markov chain model to determine the inventory shortfall distribution, which represents the number of units in inventory below a target inventory level. The shortfall distribution is independent of the target inventory level and is used to derive the optimal ordering policy for the inventory system. This work provides an efficient way to solve for the parameters in the problem.The objective of this research is to develop inventory models that minimize total expected cost per period to the manufacturer. The costs include holding cost, backorder cost, and ordering cost. A method to approximate the cost function with a quadratic least squares fitting is shown, which can be extended upon for future research. We provide numerical examples for each case in which we compare the results to a single-level reorder policy. We show that for the numerical examples tested, our policies for Case 1 and Case 2 outperform the single reorder policy by 1.6% and 2.55%, respectively.