It is a common practice to use simulation for validating different types of control and planning algorithms. However, the science of how to rigorously integrate simulation and decision models is not well understood and becomes critically important as the complexity and scale of these models increase. In our research, we have developed a methodology for integrating different types of models using a Knowledge Interchange Broker (KIB). In this paper we describe a supply-chain semiconductor application where the KIB has been used as an integral part of developing and deploying a commercial Model Predictive Control model for use in operating a multi-billion dollar supply chain. The simulation based experiments facilitated developing and validating the controller design and data automation for a real-world semiconductor manufacturing system.