A Multi-Agent Geo-Simulation Approach for the Identification of Risky Areas for Trains

The identification of risky areas along the railroad in the context of a railway system is a complex problem. A railway system is spatially and functionally distributed; its subsystems have a high degree of autonomy, and are in constant interaction with each other and with their geographic environment. In order to identify risky areas in the vicinity of rock falls zones we need to model and simulate the train behaviours in large scale geographic environments. Such a process involves coping with a variety of dynamic variables including the train characteristics, the environment properties as well as the weather conditions. The traditional mathematical and statistical modelling techniques which are usually used for the identification of risky areas do not satisfy all the requirements of such a complex process where spatial constraints are of high importance. In this context, multi-Agent geo-Simulation provides a flexible approach that can be used to easily simulate complex systems in large scale georeferenced environments. The purpose of this paper is to present Train-MAGS, an agent-based geosimulation tool which simulates train behaviours and identifies risky areas in large scale geographic environments. We show how agentbased simulation opens interesting perspectives regarding the development of new functionalities to improve risk assessment in the transportation field, more particularly for railway networks.