Traffic Signal Coordination using Termite Spatial Correlation Optimization for Oversaturated Signals

Abstract

This article proposes a novel method for solving traffic signal coordination problem under over-saturated conditions. This problem is a large combinatorial optimization problem formulated as a dynamic optimization problem. The algorithm derives optimal green times for a network consisting of 20 interconnected signals using termite spatial correlation optimization (TSCO) algorithm. The algorithm tries to do so by incorporating proper queue dissipation along with maximizing number of vehicles precessed by the network in the congestion period. The resulted green times and fitness values have been compared with those derived using Genetic Algorithm (GA) and Ant Colony Optimization (ACO). TSCO is shown to outperform GA and ACO in terms of fitness value as well as overall function evaluations.