Reinforcement learning approach for coordinated passenger inflow control of urban rail transit in peak hours

Jiang, Zhibin; Fan, Wei; Liu, Wei; Zhu, Bingqin; Gu, Jinjing

doi:10.1016/j.trc.2018.01.008

Cited by 102 publications

(38 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another example is to coordinate passenger inflow control problem on an urban rail transit line in Shanghai. In order to reduce the frequency of metro passengers being stranded and ensure public safety, Jiang et al [76] presented a reinforcement learning-based method applied to study metro passenger inflow control strategy in peak hours. These methods aim at found better optimization strategy using reinforcement learning algorithms and further optimize current traffic flow, improve the efficiency of intelligent transportation systems.…”

Section: Reinforcement Learning-based Methodsmentioning

confidence: 99%

Urban flow prediction from spatiotemporal data using machine learning: A survey

et al. 2020

View full text Add to dashboard Cite

Section: Reinforcement Learning-based Methodsmentioning

confidence: 99%

Urban flow prediction from spatiotemporal data using machine learning: A survey

et al. 2020

View full text Add to dashboard Cite

“…Xu et al [9] studied the passenger flow organization problem at subway stations under uncertain demand and developed a unified simulation-based algorithm to solve it. Jiang et al [10] developed a new reinforcement learning-based method to optimize the inflow volume during a certain period of time at crowded metro stations. In principle, the studies mentioned above aimed to relieve the passenger congestion at stations.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Therefore, increasing the capacity to satisfy booming demands is a straightforward but not always feasible in practice due to potential long construction time periods, budget limitations, right-of-way constraints, etc. [10]. The physical metro topology presents the characteristics of both small world and the power-law distribution of distance between stations [16].…”

Section: Set Of Effective Alternative Routesmentioning

confidence: 99%

“…The threshold values of 0.85 and 1 in Constraint (7) are set according to both the General Technical Specifications for Metro Vehicles (GB/T 7928-2003) [20] and the literature [3,4]. The ridership of segment(s) V k,a (n + 1) is determined by constraint (10), in which f (V i , η(n)) is a nonlinear function of the segment ridership on the route with overloaded segment(s) denoted by η(n). The specific form of function f is unknown.…”

Section: Mathematical Modelmentioning

confidence: 99%

See 1 more Smart Citation

Service-Oriented Load Balancing Approach to Alleviating Peak-Hour Congestion in a Metro Network Based on Multi-Path Accessibility

Huang

Fan

et al. 2019

Sustainability

Self Cite

View full text Add to dashboard Cite

To further improve the service quality and reduce safety risks in current congested metro systems during peak hours, this paper presents a load balancing (LB) approach so that available capacity can be utilized more effectively in order to alleviate peak hour congestion. A set of under-utilized yet effective alternative routes were searched using a deletion algorithm (DA) in order to share the passenger loads on overcrowded metro line segments. An optimization model was constructed based on an improved route generalized time utility function considering the penalties of both in-vehicle congestion and transfers. A detailed load balancing solution was generated based on the proposed algorithm. A real-world example of three overloaded metro line segments in the Shanghai metro network were selected and used to verify the feasibility and validity of the developed load balancing method. The results show that the load balancing method can effectively reduce the overcrowding situation to a great extent. Finally, two prospective inducing schemes are discussed to help implement the load balancing solution in the actual metro system in an efficient and effective manner.

show abstract

“…With millions of people choosing rail transit as their primary travel mode, the congestion occurs in the peak hour. Many measures for passenger inflow control have been taken in the Beijing and Shanghai Subway to limit the number of passengers entering the subway station and waiting at the platform [5,6]. Other measures such as pre-trip information are also used to provide travellers with help in route selection and avoiding crowded stations.…”

Section: Introductionmentioning

confidence: 99%

Understanding Daily Travel Patterns of Subway Users – An Example From the Beijing Subway

Huang

Rong

Lin

et al. 2020

PROMET

View full text Add to dashboard Cite

The daily travel patterns (DTPs) present short-term and timely characteristics of the users’ travel behaviour, and they are helpful for subway planners to better understand the travel choices and regularity of subway users (SUs) in details. While several well-known subway travel patterns have been detected, such as commuting modes and shopping modes, specific features of many patterns are still confused or omitted. Now, based on the automatic fare collection (AFC) system, a data-mining procedure to recognize DTPs of all SUs has become possible and effective. In this study, DTPs are identified by the station sequences (SSs), which are modelled from smart card transaction data of the AFC system. The data-mining procedure is applied to a large weekly sample from the Beijing Subway to understand DTPs. The results show that more than 93% SUs of the Beijing Subway travel in 7 DTPs, which are remarkably stable in share and distribution. Different DTPs have their own unique characteristics in terms of time distribution, activity duration and repeatability, which provide a wealth of information to calibrate different types of users and characterize their travel patterns.

show abstract

Reinforcement learning approach for coordinated passenger inflow control of urban rail transit in peak hours

Cited by 102 publications

References 23 publications

Urban flow prediction from spatiotemporal data using machine learning: A survey

Urban flow prediction from spatiotemporal data using machine learning: A survey

Service-Oriented Load Balancing Approach to Alleviating Peak-Hour Congestion in a Metro Network Based on Multi-Path Accessibility

Understanding Daily Travel Patterns of Subway Users – An Example From the Beijing Subway

Contact Info

Product

Resources

About