Intelligent Scheduling for Flexible Manufacturing Systems

A scheme for the scheduling of Flexible Manufacturing Systems (FMS) has been developed which integrates neural networks, parallel Monte-Carlo simulation, genetic algorithms. and machine learning. Modular neural networks are used to generate a small set of atlractive plans and schedules from a larger list of such plans and schedules. Parallel Monte-Carlo Simulation predicts rhe impact of each on the future evolution of the rnmufacturing system. Genetic algorithms are utilized to combine attractive alternatives into a single "best" decision. Induction mechanisms are used for learning and simplib the decision process in future performances. This approach forms the heart of the optimization function to be incorporated into a generic architecture for an intelligent manufacturing control ler . This paper investigates the development of a modular neural network architeciure for candidate rule selection for a FMS cell. A straightforward scheduling example illustrates the scheme capabi l i t ies including speed , adaptabi l i ty , and computational efficiency.