Model and optimize train scheduling on railways network

Abstract

In this work, a mathematical formulation of the train scheduling problem is provided as a Mixed Integer Linear Program. The MILP can be solved for easy cases, but computation time makes it impractical for more complex examples. Then a genetic algorithm is employed in the solution of the problem, with heuristic techniques to generate an initial population. The algorithm is applied to a number of problem instances producing feasible, though not optimal, solutions in less then one minute. Some improvements are suggested to obtain better results and further improve computation time.