Agent-oriented modelling for simulation of complex environments

This article addresses the application of agent-oriented modelling to composing scenarios for simulating problem domains consisting of heterogeneous entities that include humans, physical subsystems, and software components whose behaviours depend on the situation at hand. The article presents an overview of agent-oriented modelling and addresses the application of agent-oriented modelling and simulation for context-aware crisis management and military urban operations. We develop an approach for constructing vignettes of situation-aware behaviour to be further simulated by means of software agents and describe the creation of practical context-aware training scenarios. Finally, the article explores a platform currently in use for possible execution of agent-based simulations. Our approach is applicable in practice for testing typical behavioural vignettes in specific scenarios using the platform. Unique benefit of the proposed approach is its usability to observe how real subjects form their decisions to behave in certain situations.