Metro track geometry car routing problem with periodic demands: a case study in Beijing, China

Because of their curvature, age, track type and maintenance history, metro tracks should be inspected periodically during the planning horizon. Metro companies operate track geometry cars to periodically collect measurement data from the tracks by crossing the urban rail transit network. Optimizing the schedule for the track geometry car is a challenge: in addition to minimizing travelling distance, the inspection time interval of the same line should be arranged as equally as possible within the planning horizon. This article proposes a mathematical model for the track geometry car routing problem with periodic demands (TGCRP-PD). To effectively address this problem, a memetic algorithm-based metaheuristic is adopted. The proposed solution approach is applied to a real-world case. The numerical results show that the proposed method could save a dead mileage of 295.016 km (accounting for 48.882%) and greatly improve the level of service intervals.