Discrete-event modeling of heterarchical manufacturing control systems

This paper is concerned with a systematic modeling and design methodology for multi-agent control systems in manufacturing environments. The basic idea of the proposed approach is to develop discrete-event dynamical models of the single atomic decision entities (agents), and then to derive a model of the whole control network by specifying the interaction protocols between the autonomous controllers. We describe the interactions between part agents, workstation agents and transport agents. The interaction protocols are necessary to define the sequential state and the state transition mechanisms, which are the core of the DEVS-based agent model introduced by the authors.