Multi-agents approach for modelling safety interventions on a SEVESO site through virtual reality

Abstract : The new possibilities offered by virtual reality in terms of behavioural modelling and the emergence of knowledge engineering tools for risks analysis give the opportunity to envisage tools to improve training and decision making for preventing risks. In the V3S (Virtual Reality for Safe Seveso Substractors) project, we design a tool allowing storyboarding hazardous working situations on Seveso sites for risk prevention and decision making. The tool interprets a high level task and a related risk model. It is meant for a manager to help him/her to make decisions. The manager plays the scenario of an intervention and manages a tean of virtual operators (associated with autonomous agents). Depending on his/her decisions, the incurred risks are displayed in the virtual environment. In this paper, we present our project and the multi-agent architecture on which it relies.