Virtual Fusion: The Integration and Analysis of Simulation and Real Processes for Manufacturing Process Deployment

The Hybrid Process Simulation (HPS) approach presented in this dissertation provides a testing and validation platform to support the complete deployment of a manufacturing system. A process planner can start from pure simulation and progressively add real components as they arrive until the entire process is non-simulated. The HPS approach should be able to reduce ramp-up time by allowing system wide troubleshooting to occur before all process components are installed. System-wide operation is enabled by the use of component simulations that can function while interacting with the real process.;The HPS approach is based on Hardware-in-the-loop (HIL) which is a widely used testing approach for embedded systems, where real components and/or controllers are tested in closed-loop with a simulation model. This dissertation generalizes the HIL concept into HPS. An HPS is a test setup that contains at least one simulated and one actual component, but may contain many of both. A conceptual architecture is developed that separates the effect of a component from its spatial essence (volume or mass).;The conceptual architecture is applied to a small manufacturing line at the University of Michigan. A formalized method is then devised for replacing simulations with real processes and vice versa. Application of this method demonstrates how an HPS can be used to test a manufacturing system setup with multiple regions of real and simulated components.;An analysis methodology for quantitatively describing a manufacturing process during deployment, called Equivalence, is developed. The equivalence analysis put forth here has been applied to two manufacturing processes containing both real and simulated components. These examples demonstrate how equivalence analysis can be used as a tool to track and describe deployment when the HPS approach is employed.;Finally the conceptual architecture is expanded to include Autonomous Mobile Entities (AMEs) such as AGVs and human workers. This expansion includes the incorporation of simulation environments capable of simulating the laws of physics. To demonstrate an AME implementation, a game engine was used to create a three dimensional environment in which a simulated AME could move about and interact. This implementation allowed a simulated AME to control a pallet stop release in the real manufacturing process.