A DYNAMIC MANUFACTURING SYSTEM MODELLING, SIMULATION AND CONTROL FRAMEWORK

摘要

As the world market has become increasingly turbulent and unforeseen manufacturing uncertainties such as machine breakdowns, material shortages are unavoidable, the only solution for manufacturing organisations to survive is to respond rapidly to these uncertainties. The ultimate aim of this paper is to present a preliminary framework termed as Autonomous Agent Network (AAN) to help manufacturing organisations to attain rapid responsiveness in the uncertain market and manufacturing environment by achieving quantum leaps forward in manufacturing system modelling, simulation and control. AAN comprises a cluster of autonomous agents, with each agent being an integration of separate classes of objects implemented in an object-oriented manner. All manufacturing resources (i.e. machines, robot, workers, etc.) within a manufacturing shop floor are represented as agents. Agents are autonomous entities that have the ability to communicate, respond to stimuli, solve problems, make decisions and act upon decisions made. Operations of AAN involved the generation of alternative system configurations, simulation and evaluation of system configurations, selection of an optimal system configuration and generation of programs for all machines, materials handling devices and industrial controllers constituting a manufacturing system based on the selected optimal system configuration.