The Ant Colony Optimization for the Mode of Empty and Loaded Cars on Railway Freight

This paper addresses building the comprehensive model of distribution of empty and loaded cars, combining distribution method of Empty with distribution method of load. Considering demanded distribution of empty wagon while car loading and distribution of wagon flow as well as estimated situation of arrival while empty wagon distribution can reduce empty wagon-kilometres and increase empty wagon punctuality rate. In this paper, the hybrid algorithm based on Ant Colony Optimization (ACO) algorithms and Particle Swarm Optimization (PSO) algorithms is put forward to solve this linear integer programming model. Heuristic factor α and β in basal ACO are rebuilt and randomly searched by PSO, making the ACO rely on the self-adaptive search of the particles in the PSO rather than depending on artificial experience or trial and error. A satisfactory solution is given by the result of stimulant experiment. The feasibility of the proposed model and algorithm was verified with an example.