Multi Agent system Randomized in Road Traffic Multi Agent system Randomized in Road Traffic Multi Agent system Randomized in Road Traffic Multi Agent system Randomized in Road Traffic

In recent years, the rapid growth of road traffic density in the world generates a rising request for tools that can be used to analyze and control the traffic networks. Microscopic traffic simulation[1,2,3] has proven to be one of the most useful tools for analysis of various traffic systems. Thus microscopic traffic simulation has become an ever increasing field of research and development. In this paper we present a new model for microscopic traffic simulation; Traditional traffic simulation models neglect some real-life factors that need to be considered, such as the effect of random distribution in the entry of lane hence ,this paper presents the development of a concept and an associated architecture called the MAS2RT (Multi Agent system Randomized in Road Traffic), which is based on the combination of Multi-Agent Systems (MAS) [4,5], Poisson distribution in urban Traffic and a stochastic model [6,7] to capture the randomness of individual agent. The second contribution of this paper is about the internal structure of mobile agents[8], initially reacts according to the instructions of the Main agent (MA) and in the case of a lack of dynamic information, the mobile agents take decisions based on their experiences accumulated during previous interactions. The obtained results illustrate that using the randomness in the reaction of agent enhanced greatly the performance of simulation.