An Improved Ant Colony Algorithm for the Shortest Path in City’s Road Network

The shortest path between the start node and end node plays an important role in city’s road traffic network analysis system. The basic ant colony system algorithm which is a novel simulated evolutionary algorithm is studied to solve the shortest path problem. But the basic ant colony system algorithm is easy to run into the local optimum solution for shortest path. In order to solve the problem, the improved ant colony system algorithm is proposed. The improvement methods for selection strategy, local search, and information quantity modification of basic ant colony system are discussed in detail. The experiments are done in Beijing road network in China. The results of experiments show that comparing with the basic ant colony algorithm, the improved algorithm can easily converge at the global optimum for the shortest path.