1 DYNAMIC MODELING AND STABILITY OF THE RECONFIGURATION OF MANUFACTURING SYSTEMS Farshid Maghami Asl 1 , A. Galip Ulsoy 2 and Yoram Koren 3 Department of Mechanical Engineering and Applied Mechanics University of Michigan, Ann Arbor, Michigan 48109-2125, USA ABSTRACT Reconfigurable Manufacturing Systems (RMS) are designed with certain key characteristics (e.g. modularity) such that they can be upgraded in response to market demands. The upgrading of the system can be in terms of either capacity (volume of parts that can be produced on the line) or functionality (number of different types of parts that can be produced on the line). In other words, an RMS can provide exactly the capacity and functionality needed, exactly when needed. The complex nature of RMS and the need for a reconfiguration policy requires the development of a model to assist manufacturing engineers to implement the optimum reconfiguration policy. A fluid dynamic system analogy for RMS, using a control-theoretic approach, is developed, and several concepts and ideas of RMS such as reconfiguration and its relationship to scheduling are demonstrated using this analogy. This analogy is the basis for a dynamic model which characterizes the reconfiguration policy and the production scheduling of an RMS. We also develop and analyze a useful simplified dynamic model, by ignoring the effects of scheduling on its dynamics, of a production system whose capacity and/or functionality can change over time. It is shown, based on the simplified model that there are conditions for stability that depend on the time delay (T) associated with implementing the changes in the production system, and on a gain (K) reflecting the rate of reconfiguration. A measure of performance is developed to quantify the potential benefits of RMS over traditional dedicated or flexible manufacturing systems that have fixed capacity and/or functionality. The results of this analysis show that for changes in capacity and/or functionality to be effective, the product of the delay (T) in responding to market demand, and the rate of reconfiguration (K), must be bounded. The results are applied to study a reconfigurable manufacturing scenario, and the advantages of implementing RMS are presented. Japan-USA Symposium on Flexible Automation, Ann Arbor, MI July 2000 1 Graduate Research Assistant, Engineering Research Center for Reconfigurable Machining Systems 2 Professor and Deputy Director, Engineering Research Center for Reconfigurable Machining Systems 3 Professor and Director, Engineering Research Center for Reconfigurable Machining Systems