| Effective production planning and control, combined with effective shop-floor level scheduling, enables manufacturing with a Just-In-Time philosophy while managing the available resources, release of orders to meet due dates, and work-in-progress inventory.;Previous work at the University of Wisconsin-Madison has shown that selecting the proper continuous variables, discrete variables and control laws transforms certain distributed scheduling problems from the discrete sequencing domain to the dynamical systems domain. This past work has shown that there is an opportunity to expand this approach to encompass higher levels of production planning and control, and to improve understanding of the dynamics of these traditionally multilevel systems, particularly when they have a distributed implementation.;Distribution decision-making in modern manufacturing in the global economy can reduce the effectiveness of such controls, and there is a void in the understanding of such distributed systems and a lack of confidence in their dynamics. The emphasis of this research therefore was on modeling and understanding the dynamics of distributed manufacturing control systems with a focus on integrating dynamic scheduling with a production planning and control system. A distributed order-driven arrival-time controller, which was developed in previous work, was modified to accommodate adjustments in capacity and integrated with control systems for both WIP and due date deviation regulation. A dynamic model was developed and used to predict system performance, stability, and responses to production disturbances.;An order due date deviation regulation topology is presented for workstations that dynamically adjust their capacity and order release times. The relationship between due date deviations and workstation capacity is shown to be nonlinear and time varying, and a method is presented for characterizing the relationship quantitatively in real time and using this information in adaptive capacity adjustment control laws that maintain favorable dynamic behavior in the presence of the nonlinearities. It is anticipated that this work will build confidence in distributed manufacturing control systems and will contribute to the development of better tools for understanding the dynamics of other distributed decision-making systems such as those used in e-design and e-commerce. |
