Faced with the increasing of natural, industrial and terrorist's disasters and crisis, the organisations must improve their practices to increase their resilience. In this aim, the objective of this work is, firstly the increase of knowledge on these situations, and on the other hand, the development of a decision making aid process to anticipate and forecast a large scale crisis for the actors of the civil protection. This work is composed of three parts. The first focus on the improvement of knowledge on the base of real cases and lessons learnt observations and a state of the art. The second part proposes a definition and a systemic modelling of the large scale crisis. It defines, in theory, the decision making aid process based on two methods that are the analytical hierarchical process of Saaty and the artificial network. Finally, the third part is allotted to the application of the two approaches in the aim to evaluate the crisis potential of eighteen lessons learnt.
