Modeling and simulation of multi-agent systems: Principle influence / reaction

This paper presents a state of the art modeling techniques, namely the macroscopic model, mesoscopic and microscopic. And we will see that in order to solve problems related to space and time encountered in the above models, a new scale model called hybrid appeared Its purpose is to take advantage of conventional models, by adopting the model of the phenomenon. The simulation of multi-agent systems based on the direct representation of behaviors, actions and interactions of a set of autonomous entities operating in a common environment. In this context, the global dynamics of a system at the macroscopic level is seen as the result of the outcome of dynamic interactions that take place at the microscopic level. One of the problems associated with this approach is the link between these two levels of modeling, which is fundamental. However, deduct the macroscopic dynamics of a system based on the dynamics of microscopic level raises many technical and conceptual problems. The model Influence / Reaction of Ferber and Müller will be discussed in detail in this article we will emphasize its shortcomings, and present later the IRM4S (An Influence Reaction Model for Simulation) model that clearly explains the principleand which allows to articulate modeling micro level with the macro-level.