Multi-agent systems based distributed intelligent simulation - A case study

Abstract

AL The actual discrete-event simulation scenario reveals a systematic demand for more sophisticated and powerful computational tools in order to full the increasing of problem solving complexity. Computer network technology provides the computational power needed. The combination of Distributed Articial Intelligence technics with that of Multi Agent Systems presents a very interesting perspective to design and develop such kind of environments Multiple computing paradigms combination integration of diferent skills and knowledge sources new forms to speed up process execution and more robust system are some of the practical gains of a simulation environment that integrates knowledge and experience from the referred fields. A computational framework based on a shared memory model and on a message passing perspective to distributed and concurrent computing The attainment of maximum speed up is the goal and reward for this new generation of simulation technology. Human expertise and knowledge are emulated through agents A quick review of the distributed computing environment will be presented the approach being applied to model an Air-Conditioneds distribution net