tailieunhanh - Disordered Cellular automata traffic flow models

In this paper, we extend the Nagel-Schreckenberg (NaSch) model by introducing disordered acceleration and deceleration terms. The disorder leads to segregated states where the flow is constant at intermediate densities for high values of breaking probability p. Within the model we present a density wave behavior appears below a critical value of p. Such result was found in car following models with an optimal velocity. The behavior of the gap distribution shows that the traffic exhibits a self organized criticality for high values of p and random deceleration. | M. J. CONDENSED MATTER VOLUME 5 NUMBER 2 June 2004 Disordered Cellular automata traffic flow models and LMPHE Departement de physique Faculté des sciences Agdal Rabat Morocco. In this paper we extend the Nagel-Schreckenberg NaSch model by introducing disordered acceleration and deceleration terms. The disorder leads to segregated states where the flow is constant at intermediate densities for high values of breaking probability p. Within the model we present a density wave behavior appears below a critical value of p. Such result was found in car following models with an optimal velocity. The behavior of the gap distribution shows that the traffic exhibits a self organized criticality for high values ofp and random deceleration. PACS numbers . w Introduction Recently traffic problems have attracted considerable attention due to the fact that traffic behavior is important in our life. In recent years many non-equilibrium systems have been modeled 1 2 as systems of interacting particles driven far from equilibrium. A special class of such models includes for example car- following models cellular automaton models hydrodynamic models3 and gaz kinetic models4 . For a detailed study of all above approaches and theories we refer the reader to the review article ref. 5. For their simplicities cellular automaton CA models have been used to analyze traffic problems. The NaSch model and its slow to start variant 6 is the simplest one that reproduces the mean features of the real traffic. It is a stochastic CA model based on some pertinent rules. The model was intensively studied using both analytical and numerical methods 7 . In this work we suggest and study numerically an extension of the NaSch 8 9 model that presents many features of real traffic. In our model unlike in the NaSch model the particles cars may accelerate or decelerate randomly more than one unit in single update step. We mention that such phenomenon were