International Journal of Railway Research
Iran University of Science & Technology
Sat, Jul 19, 2025
[Archive]
Volume 12, Issue 2 (7-2025)
IJRARE 2025, 12(2): 1-12 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
sedigh S, Shafahi Y. The Impact of Climate Conditions on the Geometry of Railway Tracks in Iran: A Deep Learning Approach. IJRARE 2025; 12 (2) :1-12
URL: http://ijrare.iust.ac.ir/article-1-376-en.html
URL: http://ijrare.iust.ac.ir/article-1-376-en.html
The Impact of Climate Conditions on the Geometry of Railway Tracks in Iran: A Deep Learning Approach
epartment of Civil Engineering, Sharif University of Technology, Tehran, Iran
Abstract: (264 Views)
Climate is one of the most important factors in railway infrastructure, and future changes in climate and extreme weather conditions will increase the risks concerning its stability and performance. Understanding and mitigating these risks is vital for ensuring the longevity and safety of railway networks. The research work presented here examines the association of climatic conditions with the geometric degradation of railway lines in Iran using Deep Learning models.
The research is based on the geometric data obtained from track measurement machines over 14 years, encompassing track gauge, profile, and twist. These data were matched with meteorological records such as temperature, humidity, rainfall, and wind speed for different regions of Iran. By combining these datasets, deep learning models are developed to analyze and predict the pattern of track degradation resulting from various climatic conditions.
The added value of using meteorological data in training predictive models is assessed by comparing the performance of models trained with and without meteorological data. More precisely, one model predicts track degradation without considering meteorological data, while the other includes climatic information. This comparison allows for an evaluation of the effectiveness of weather data in improving the accuracy of track degradation predictions.
This research is, therefore, likely to contribute to critical knowledge of how climatic variables influence railway infrastructure, allowing for more realistic forecasts of track life and better planning for maintenance schedules.
Given the challenges associated with climate change, the present study addresses the call to develop more resilient railway infrastructure for operational safety across varied climatic conditions.
The research is based on the geometric data obtained from track measurement machines over 14 years, encompassing track gauge, profile, and twist. These data were matched with meteorological records such as temperature, humidity, rainfall, and wind speed for different regions of Iran. By combining these datasets, deep learning models are developed to analyze and predict the pattern of track degradation resulting from various climatic conditions.
The added value of using meteorological data in training predictive models is assessed by comparing the performance of models trained with and without meteorological data. More precisely, one model predicts track degradation without considering meteorological data, while the other includes climatic information. This comparison allows for an evaluation of the effectiveness of weather data in improving the accuracy of track degradation predictions.
This research is, therefore, likely to contribute to critical knowledge of how climatic variables influence railway infrastructure, allowing for more realistic forecasts of track life and better planning for maintenance schedules.
Given the challenges associated with climate change, the present study addresses the call to develop more resilient railway infrastructure for operational safety across varied climatic conditions.
Type of Study: Research |
Subject:
Railway Transportation
Send email to the article author
| Rights and permissions | |
 | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
