When urban rail transit is faced with a large number of commuter passengers during peak periods, passengers are often waiting for the next train because the subway is running at full load, which causes delays to the overall travel time of passengers. The calculation and prediction of the congestion delay in subway stations can guide the operation department and passengers to make better planning and selection. In this paper, we use a new method based on deep learning technology to evaluate the congestion delay of subway stations. Firstly, we use automatic fare collection (AFC) system data to evaluate the congestion delays of stations. Then, we use a convolutional long short-term memory (Conv-LSTM) network to extract spatial and temporal characteristics to solve the short-term prediction problem of the subway congestion delay in the network structure. The spatiotemporal variables include inbound passenger flow, outbound passenger flow, number of passengers delayed, and average delay time. As a spatiotemporal sequence, the input and prediction targets are both spatiotemporal three-dimensional tensors in the end-to-end training model. The effectiveness of the method is verified by a case study of the Chongqing Rail Transit. Experimental results show that Conv-LSTM is better than the benchmark models in capturing spatial and temporal correlation.


    Access

    Download


    Export, share and cite



    Title :

    A Deep Learning Model with Conv-LSTM Networks for Subway Passenger Congestion Delay Prediction


    Contributors:
    Wei Chen (author) / Zongping Li (author) / Can Liu (author) / Yi Ai (author)


    Publication date :

    2021




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    Unknown





    Instant traffic flow prediction method based on Conv-LSTM

    WU LIJUN | European Patent Office | 2023

    Free access

    A novel hybrid deep learning model with ARIMA Conv-LSTM networks and shuffle attention layer for short-term traffic flow prediction

    Sattarzadeh, Ali Reza / Kutadinata, Ronny J. / Pathirana, Pubudu N. et al. | Taylor & Francis Verlag | 2025

    Free access

    A Semi-Conv-Transformer Model for Inflow Prediction of Newly Expanding Subway Lines

    Mo, Yue / Zhang, Jinlei / Wang, Chengcheng et al. | IEEE | 2025


    Conv-Bi-LSTM model-based ship heaving motion prediction system and prediction method

    ZHANG YA / FAN SHIWEI / PING BAOJIN et al. | European Patent Office | 2023

    Free access