ABSTRACT
Distributed supervisory control systems consist of a group of agents (humans or computers) that control a complex, dynamic, event-driven system (e.g., nuclear power plants, satellite ground control). The control is distributed in the sense that each of the agents has a particular responsibility in the control process. For example, agents may be responsible for controlling a single subsystem or for managing and coordinating other agents. The control is supervisory in that the agents supervise a highly automated process; their major duties typically include system configuration, monitoring, occasional control actions to fine-tune the system, the execution of standard operating procedures, scheduling and preplanning of future activities, and diagnosis and compensation for abnormal system conditions (e.g., Hollnagel, Mancini, & Woods, 1988; Jones & Mitchell, 1995; Sheridan & Johannsen, 1976).
