Traffic flow estimation using ant colony optimization algorithms

Simulation and optimization of traffic flows in a city or province allow the implementation of correct developing strategies and help the decision making process when using and distributing resources such as mass transit. This estimation can be modeled as a bifurcated multi-commodity network flow problem, where the general flow distribution is dictated by Wardrop's principles. In this paper two different Ant Colony Optimization algorithms are presented for solving this problem. The proposed algorithms are tested with real-life traffic demand in the Havana city. The obtained results are compared to those provided by classical algorithms, showing that the new ant colony algorithms provide good results as well as low running times.