Some firms make all their products to order while others make them to stock. There are a number of firms that maintain a middle ground, where some items are made to stock and others are made to order. This paper was motivated by a consumer product company faced with the decision about which items to make to stock and which ones to make to order, and the inventory and production policy for the make-to-stock items. The production environment is characterized by multiple items, setup times between the production of consecutive items, limited capacity, and congestion effects. In such an environment, making an item to order reduces inventory costs for that item, but might increase the lot size and inventory costs for the items made to stock. Also, lead times increase because of congestion effects, resulting in higher safety stocks for make-to-stock items and lower service levels for make-to-order items, thus leading to a complex trade-off. We develop a nonlinear, integer programming formulation of the problem. We present an efficient heuristic to solve the problem, which was motivated by key results for a special case of the problem without congestion effects that can be solved optimally. We also develop a lower bound to evaluate the performance of the heuristic. A computational study indicates that the heuristic performs well. We discuss the application of the model in a large firm and the resulting insights. We also provide insights into the impact of various problem parameters on the make-to-order versus make-to-stock decisions using a computational study. In particular, we find that the average number of setups of an item selected for make-to-stock production is always less than half the average number of setups of the item if it were to be made to order. Also, factors other than an item's demand, such as its setup time, processing time, and unit holding cost, impact the make-to-order versus make-to-stock decision.
