Computational Technology Resources - CCC

T. Fukuda, S. Takahashi, H. Nakamura, "An Examination of Machine Learning Methods for Predicting Station Departure Delay Time Based on Historical Train Traffic Records", in J. Pombo, (Editor), "Proceedings of the Sixth International Conference on Railway Technology: Research, Development and Maintenance", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 7, Paper 10.9, 2024, doi:10.4203/ccc.7.10.9

Keywords: train, machine learning, simulator, delay countermeasures, train scheduling, robustness.

Abstract

In the urban railways of the metropolitan area, train delays during morning rush hours have become a significant issue. This problem isn't merely caused by large, sudden delays, but also by an accumulation of minor delays, which can escalate into substantial setbacks. This presents a disadvantage not only for the operators managing the train services but also for the users. Recent years have seen active reporting of studies predicting delay times. However, these studies do not take into account the accumulative nature of delays because the data input to the learning machine is only for stations behind the target station. Therefore, this paper aims to predict the delay time that occurs when each train departs from a station. To do so, we constructed a network that takes into account the propagation of delays by using data from stations in front of the target station as input. We will provide an overview of the network we constructed and report on its predictive accuracy.

download the full-text of this paper (PDF, 11 pages, 928 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 7 PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE Edited by: J. Pombo Paper 10.9 An Examination of Machine Learning Methods for Predicting Station Departure Delay Time Based on Historical Train Traffic Records T. Fukuda, S. Takahashi and H. Nakamura College of Science and Technology, Nihon University, Japan doi:10.4203/ccc.7.10.9 Full Bibliographic Reference for this paper T. Fukuda, S. Takahashi, H. Nakamura, "An Examination of Machine Learning Methods for Predicting Station Departure Delay Time Based on Historical Train Traffic Records", in J. Pombo, (Editor), "Proceedings of the Sixth International Conference on Railway Technology: Research, Development and Maintenance", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 7, Paper 10.9, 2024, doi:10.4203/ccc.7.10.9 Keywords: train, machine learning, simulator, delay countermeasures, train scheduling, robustness. Abstract In the urban railways of the metropolitan area, train delays during morning rush hours have become a significant issue. This problem isn't merely caused by large, sudden delays, but also by an accumulation of minor delays, which can escalate into substantial setbacks. This presents a disadvantage not only for the operators managing the train services but also for the users. Recent years have seen active reporting of studies predicting delay times. However, these studies do not take into account the accumulative nature of delays because the data input to the learning machine is only for stations behind the target station. Therefore, this paper aims to predict the delay time that occurs when each train departs from a station. To do so, we constructed a network that takes into account the propagation of delays by using data from stations in front of the target station as input. We will provide an overview of the network we constructed and report on its predictive accuracy. download the full-text of this paper (PDF, 11 pages, 928 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions