Streamlined Hierarchical Topology Network-Based Passenger Flow Assignment of Urban Rail Transit. Issue 8 (August 2022)

• Record Type: Journal Article
• Authors: Hao, Yuanyuan; Bingfeng, Si; Zhao, Chunliang
• Is Part Of: Transportation research record. Volume 2676:Issue 8(2022)
• Page Start: 683
• Page End: 696
• Publication Date: 2022-08
• Subjects: planning and analysis -- algorithms -- mathematical modeling -- network assignment -- path finding -- rail; Transportation -- Periodicals; Roads; Transport -- Périodiques; Routes -- Périodiques; Routes -- Conception et construction -- Périodiques; Roads; Transportation; 388.05
• DOI: 10.1177/03611981221084689
• Languages: English
• ISSNs: 0361-1981

Abstract

In urban rail transit systems, passenger flow assignment is vital for administrators and designers. Many approaches using evaluation indicators seek to improve assignment accuracy. However, such approaches suffer from the inefficiency in traffic flow computation, especially in large-scale networks. K-shortest path searching is an essential component in flow assignment; its complexity depends on the network scale. This study proposes a simplified framework with a bi-layer network (BL-NSF) to find the K-shortest path as early as possible to improve the speed of flow assignment while meeting accuracy requirements. First, a bi-layer transformation operator is developed. Specifically, a first-layer transformation operator employs the station extraction function to simplify the type of stations. The network is, therefore, first reduced since non-transfer stations are extracted. Second, a second-layer transformation operator with a path filter function is designed to eliminate the non-effective connection structure. Thus, the complexity of the network is further reduced, and the streamline structure of the topology rail network is constructed. Sequentially, we search the K-shortest path by executing four designed matching mechanisms and a modified Yen's algorithm, a traditional algorithm used in the existing transit system. After that, we perform passenger flow calculations for stations, links, and lines. Finally, a series of experiments is run on the test network to verify the effectiveness and efficiency of BL-NSF on the flow assignment. The results demonstrate that the proposed BL-NSF performs better and is more competitive than the state-of-the-art algorithms.