A turn-back track constraint train scheduling algorithm on a multi-interval rail transit line

With the rapid increase of passenger volume, the interval of some rail transit lines in China has reached a minimum value, which makes turn-back capacity one of the main capacity-limiting factors instead of section headway, particularly during the peak hours on workdays. A general simulation modeling framework and algorithm in which the divide and conquer rule is adopted for train scheduling problems on a multi-interval rail transit line with turn-back track constraint is proposed in the paper, and the objective is to obtain a feasible timetable based on passenger demand at different time periods, line capacity, number of rolling stocks, location of depots, train routing, turn-back type. Lastly, a new version of the TPM software is programmed by the proposed framework and algorithm, and Shanghai Rail Transit Line 2 is presented to demonstrate. The result shows that the proposed algorithm performs very efficiently, and a feasible timetable with a five-interval period, 535 trains and one turn-back track constraint at GLR station is generated in 15 seconds by TPM.