Optimization algorithm for dynamic multi-agent job routing

Abstract

Current research in multi-agent heterarchical control for holonic systems is usually focused in real-time scheduling algorithms, where agents explore the routing or process sequencing flexibility in real-time. In this paper we investigate the impact of the dynamic job routing andjob sequencing decisions on the overall optimization of the system's performance. An approach to the optimization of local decisions to assure global optimization is developed within the fvamework of a Neural Collective Intelligence (NECOIN). Reinforcement learning (RL) algorithms are used at the local level, while generalization of Q-neural algorithm is used to optimize the global behaviour. A simulation test bed for the evaluation of such types of multi-agent control architectures for holonic manufacturing systems integrating discrete-event simulation facilities is implemented over JADE agent platform. Performance results of the simulation experiments are presented and discussed. © 2005 by International Federation for Information Processing.