A train dispatching optimization for classification yards by genetic algorithm

This article formulates an optimization model of train dispatching problem for double-direction classification yards, whose goal is to provide a method for choosing an appropriate receiving or departing system for each train on real time horizon. The model constrained by available resources is based on systematic analysis of the operating procedure and process in the yard. The formulation developed here is a mixed 0-1 programming problem and Genetic Algorithm is used to obtain a near optimal solution. The implementation of the algorithm applied to a real case in China is also introduced.