Enhancing Intelligent Mobility: LSTM Neural Networks for Short-Term Passenger Flow Prediction in Rail Transit

Jiajun Zhang, Jingrong Zhang

2024

Abstract

In Urban Rail Systems (URS), traffic flow prediction has a long history. However, due to the inherent high non-linearity and randomness of transportation systems, it remains a challenging issue. Based on the passenger flow data from subway stations in Seoul, South Korea, this study aims to conduct short-term passenger flow predictions within the Seoul Metropolitan Subway system in South Korea using the Long Short-Term Memory (LSTM) network model, thereby verifying the effectiveness and accuracy of the LSTM model in urban subway passenger flow prediction. The model aids in facilitating early safety warnings and evacuations for passenger flow. According to the results, the LSTM model is more accurate for short-term passenger flow prediction in a high traffic station ("Geongdeok Station") and for a stable station ("Jamwon Station"). Compared to traditional time-series prediction models, LSTM shows superior forecasting capabilities. The findings and methodology in this study could serve as references and lessons for other researchers in similar fields.

Paper Citation

in Harvard Style

Zhang J. and Zhang J. (2024). Enhancing Intelligent Mobility: LSTM Neural Networks for Short-Term Passenger Flow Prediction in Rail Transit. In Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-690-3, SciTePress, pages 410-415. DOI: 10.5220/0012797600004547

in Bibtex Style

@conference{icdse24,
author={Jiajun Zhang and Jingrong Zhang},
title={Enhancing Intelligent Mobility: LSTM Neural Networks for Short-Term Passenger Flow Prediction in Rail Transit},
booktitle={Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2024},
pages={410-415},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012797600004547},
isbn={978-989-758-690-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - Enhancing Intelligent Mobility: LSTM Neural Networks for Short-Term Passenger Flow Prediction in Rail Transit
SN - 978-989-758-690-3
AU - Zhang J.
AU - Zhang J.
PY - 2024
SP - 410
EP - 415
DO - 10.5220/0012797600004547
PB - SciTePress