A Smart Path Recommendation Method for Metro Systems with Passenger Preferences

Abstract

Passenger travel paths in metro networks have become more diversified with the development of network structures and the complexity of train schedules. Nowadays, passengers may have more than one alternative path in an OD (Origin-Destination) pair. In order to provide high-quality service to passengers, this paper proposes a smart path finding method to recommend fast and comfortable routes to passengers. By including the structure of the metro network as a two-dimensional plane and time as the third dimension, the space-time range of passenger activities is constructed. The accessible transfer stations for arrival are firstly computing by forward searching from O station, and the accessible transfer stations for departure are then identified by backward searching from D station. The intersection of the two feasible trajectories is regarded as the feasible path set of passengers. The path is recommended to passengers using generalized path impedance with the consideration of the situation that passengers have different perceptions of travel time, transfer penalty, and congestion tolerance. Finally, the proposed method is validated with the data of Shenzhen Metro under different traffic demand conditions, including the uncongested condition, the condition with congestion, the last train condition, and the condition with train delays, respectively. The time complexity of the proposed algorithm is also compared with that of some conventional algorithms. The results show that the proposed algorithm has lower algorithm complexity and is more suitable for the increasingly large-scale metro network.