Optimization of an MRT train schedule: reducing maximum traction power by using genetic algorithms

Because of the traction power load characteristics of mass rapid transit (MRT) systems, a significant difference exists between the peak and valley of the traction power load curve. Furthermore, an extremely high traction power peak occurs if numerous trains are accelerating simultaneously. Careful train scheduling can avoid the simultaneous acceleration of too many trains, thus reducing maximum traction power. This paper employs genetic algorithms to optimize train scheduling and uses the Kaohsiung MRT system as an example. Simulation results show that the proposed method can significantly reduce the maximum traction power.