| Supply chain management (SCM) is concerned with the efficient movement of goods through a network of suppliers and retailers. Effective SCM represents a crucial imperative for all modern, global enterprises. This dissertation discusses a series of fundamental research problems that lead to greater insight into the supply chain problem space and the development of next-generation control-oriented decision policies for supply chain management.;Among the important control concepts illustrated are the modeling of supply chain dynamics using fluid analogies, the benefits of multi-degree-of-freedom feedback-feedforward control, and the application of Internal Model Control (IMC) and Model Predictive Control (MPC). The IMC and MPC decision policies are formulated for optimality with respect to their underlying nominal linear models, but need to be tuned under stochastic, nonlinear real-world conditions. The effective use of simulation-based optimization, in conjunction with Simultaneous Perturbation Stochastic Approximation, serves to enhance the performance of these decision policies.;Forecasting highly uncertain demand signals is important for successfully managing inventory in semiconductor supply chains. This dissertation presents a control-relevant approach to the problem that allows the supply chain planner to generate demand forecasts that minimize inventory deviation, starts change variance, or their weighted combination when incorporated in an MPC decision policy. Nonlinear system identification techniques are applied to data from a discrete event simulation, resulting in a continuity-based fluid model that accurately captures the dynamics of the manufacturing system in question. The resulting models provide a more accurate basis for the analysis and design of tactical decision policies.;The industrial outcomes of this dissertation are expected to yield greater customer satisfaction, increased financial benefits, and decreased environmental impacts. |
