Using Stochastic Petri Nets for Level-Crossing Collision Risk Assessment

Level crossings (LCs) are identified as critical security points in both road and rail infrastructures. Statistics show that more than 300 people are killed every year in Europe in more than 1200 accidents occurring at LCs. In this paper, we first propose a global model involving both rail and road traffic in the LC area. This model is obtained by a progressive integration of elementary models that we developed, each of which describes the behavior of a part in the whole LC environment. We are more precisely interested in a particular phenomenon that may cause collisions at LCs and corresponds to the accumulation of vehicles' waiting queues at the LC exit zone. As a notation, we use stochastic Petri nets (SPNs) in such a way as to precisely reflect the system's dynamics. Second, the simulation of the global system behavior is performed in light of the behavioral model while adopting the Monte Carlo principle. The TimeNet tool is used as a simulator that allows the monitoring of risky situations. To qualitatively and quantitatively assess the effect of various factors on the risk level, setup tasks are undertaken. Finally, the simulation results are analyzed and interpreted. This analysis makes it possible to consider some solutions to reduce the incurred risk.