The aim of this paper is to present the use of particle method to evaluate the stress modes of a deploying Airbag. Starting from its folded condition.

The innovation is in the gas modelization which is assumed to be made off a large number of particles. As that number is bounded by memory and CPU performance, it is clear that the particles are not identified with the molecules. They are instead abstract building blocks, whose dynamics are chosen to reflect the macroscopic behaviour of the gas.

The natural time scales of the molecular system are of the order of 10^-8 seconds, the typical time interval between collisions. For obvious reasons, the time step of the simulation should integrate the effects of many molecular events.

The approach consists of replacing the molecular content of the system by a different «microscopic» description, whose coarse-grained space~time behaviour approximates the macroscopic dynamics.

The main idea is to consider the gas as an hybrid system containing a fluid in the classical sense (obeying fluid dynamic equations) and particles in interaction with it. Both elements correspond to the same physical system, but they represent different aspects of it. The fluid represents the part which is in thermal equilibrium, while the particles handle its turbulent aspect.

As the experience with this method is innexistant, it is needed to fine-tune the model in order to build up a predictive capability.

Therefore, it is necessary to proceed by step, here the simulation will be compared to experiment results for a flat unfolded Eurobag simply lining on a flat surface.