Abstract

Here we study the possibility of characterizing city transportation in analogy to diffusing particles. To recreate the dynamics of the vehicles in a city, we use a cellular automata model in a road network. The agents represent vehicles that follow consecutive routes between origin and destination, resembling the dynamics of taxis, deliveries, carpooling, and rideshare vehicles. We calculate the mean velocity and the diffusion coefficient through the statistical analysis of the parametric curves produced by car movements. We found that a power law relationship closely relates both quantities as in kinetic theory but with a different exponent than those found in Brownian motion theory. We close the paper by discussing the possibility of using the Diffusion coefficient to characterize the city, as it is traditionally done with the mean speed and the flux rate; and how to calculate this quantity in a smart city.