Real-Time Simulation and Control of Large Scale Distributed Discrete Event Systems

Due to the ever-increasing computational capabilities and the increasing use of automated, distributed large-scale systems, intelligent controllers are being proposed and designed. Much of the discrete-event automated control research has been done in the context of manufacturing plants and as such this paper will presents the results in the context of manufacturing systems. There is a large amount of research in the area of simulation of manufacturing systems, scheduling and some research in the control of automated manufacturing systems. Even intelligent controllers for manufacturing systems have been looked at by some. To this date there has been no realization of a general-purpose intelligent controller for manufacturing systems or any other large scale discrete-event system. Furthermore the current research efforts focusing on this problem are not considering the fact that the controller must be physically distributed among many computers in order to achieve the computer power necessary to control these large-scale complex systems. Since todays large complex systems tend to be physically distributed over potentially thousands of miles, the control structure must be distributed as well. This results in a serious implication; the model used in the controller actually consists of a set of independent models. Very little research has focused on the design of a controller whose model is distributed. Presented in this paper is a controller that is designed to control the real-time discrete-event aspects of a large scale distributed system using manufacturing as a base problem. The paper concludes by presenting an example of a flexible manufacturing system being controlled by this software.