Cellular automata and its application to the modeling of vehicular traffic in the city of Caracas

In this paper an emergent microscopic traffic model based on a cellular automaton is presented. The model is part of a vehicular traffic study recently initiated in the city of Caracas in Venezuela. The proposed simulation model is an extension of the Nagel and Schreckenberg model for identical vehicles incorporating several important features: Velocities of cars are picked from a Gaussian distribution to take into account that not every car driver honors velocity limits. The model is validated by fitting measured normalized average vehicle flows by means of an iterative unconstrained optimization algorithm. For this purpose mean and variance of the velocity distribution are considered as optimization parameters together with other model parameters. Objective Functions quantifying the mean square deviations of the differences of measured and simulated normalized averages flows, are defined. The results show that the proposed simulation models reproduce satisfactorily the general features of empirical flow measurements.