Production control and capacity configuration policies are critical to a manufacturing firm for effective inventory control. In the first part of this dissertation, a Dynamic Programming model and a solution algorithm are developed to obtain an optimal (near-optimal) production control policy. The solution algorithm is able to produce an extremely good policy under mild conditions, but is applicable only to problems with a limited number of products. For problems involving a large number of products, a heuristic algorithm based on a decomposition/aggregation scheme is then proposed. This algorithm overcomes the computational difficulty typically associated with Dynamic Programming problems with a large number of state dimensions. Computational test results are reported to show the performance of the policy generated by the heuristic algorithm. In the second part of the dissertation, the production lead time and operational cost performance of two capacity configurations are analyzed. Models are developed for each configuration to determine the amount of capacity which minimizes the total capacity acquisition and operational costs, including the inventory cost. Computational test results are presented to study the impact of problem characteristics on the superiority of each configuration.