A system of emotional agents for decision-support

To facilitate the study of numerous phenomena, we use computer simulation tools. However, simulation of human behavior is still a challenge for both computer and human sciences. Multi-agent systems have already proved their efficiency to simulate complex interacting systems. In this context, we have worked on the simulation of human groups to study the emergence of behavioral patterns, mixing techniques of artificial intelligence, multi-agent systems and psychological sciences. This paper presents a new model of emotional agent, the Emotion, Feeling, Temperament agent (EFT). Based on BDI architecture, our model integrates the OCC emotional model and the PerformanSe behavioral model. We also present a concrete implementation of our model: the simulation of brain-damaged people's behavior on a production line. Then, we describe the modeling of the interactions of this system with AgentUML. The final goal of our work is to produce simulation data than can be enhanced through a KDD process.