Agent-Oriented Modelling of Real-Time Systems for Critical Events

In this paper we outline a task-centered agent-oriented approach to modelling (or software engineering) of intelligent real-time systems for critical events like environmental disasters (e.g., bush fires, earthquakes), alarm processing in critical and emergency situations in power systems, and in war situations involving detection and destruction of enemy planes by fighter pilots. We model the user (e.g., operator or fighter pilot) at two distinct levels. These include the problem solving or the task level and performance optimisation level. Alarm processing is used as an application to illustrate the modelling at the task and performance optimisation levels respectively