Rockfall is a major risk in mountainous zones. In the field of falling rocks trajectory simulations, many scientific problems related to modelling the bouncing phase remain to be solved. The large variability of slope properties and boulder removal conditions pleads for a stochastic modelling of these trajectories. In particular, the rebound of falling rocks has to be modelled as a random process (Bourrier et al., 2009a). The aim of this study is to define a stochastic impact model that can capture the variability of the bouncing phenomenon on coarse soils, in particular. For this purpose, a numerical approach based on the Discrete Element Method (DEM) is used to investigate the impact of a boulder on a coarse granular soil in 2D. This model is used to perform an intensive simulations campaign. The analysis of results from this campaign is carried out using advanced statistical methods leading to the definition of a stochastic impact model. The stochastic impact model is finally integrated in a falling rocks trajectories simulation code and the code developed is evaluated by comparing real-scale experimental results to rockfall trajectory simulations.