Asset Details
Do you wish to reserve the book?
A Machine Learning Framework for Predicting Failures in Rail Infrastructure Assets
Hey, we have placed the reservation for you!
By the way, why not check out events that you can attend while you pick your title.
Oops! Something went wrong.
Looks like we were not able to place the reservation. Kindly try again later.
Are you sure you want to remove the book from the shelf?
A Machine Learning Framework for Predicting Failures in Rail Infrastructure Assets
Do you wish to request the book?
A Machine Learning Framework for Predicting Failures in Rail Infrastructure Assets
Please be aware that the book you have requested cannot be checked out. If you would like to checkout this book, you can reserve another copy
We have requested the book for you!
Your request is successful and it will be processed during the Library working hours. Please check the status of your request in My Requests.
Oops! Something went wrong.
Looks like we were not able to place your request. Kindly try again later.
Dissertation
A Machine Learning Framework for Predicting Failures in Rail Infrastructure Assets
2025
Overview
Infrastructure safety is crucial for the rail industry, with signal functionality and track integrity being among essential components. This thesis presents a machine learning framework to predict failures in rail infrastructure assets, focusing on two critical areas: urban rail transit signal failures and broken rails in commuter rail systems. Integrating historical failure data, maintenance data, and track condition data, and operational data, the proposed framework applies machine learning models to identify high-risk locations and predict rail asset failures. Because rail infrastructure asset failures are relatively rare, imbalanced data mining techniques such as SMOTE, ADASYN, and random resampling are also employed to improve predictive accuracy.In the first case study, our model achieves an AUC of 75% and demonstrates the ability to identify approximately one-third of rail signal failures by focusing on 10% of signal locations on the network within the one-month prediction period. Our second case study focused on commuter rail segments, in which our model gives an AUC of 74% and 71% overall accuracy. The results show the potential of this framework to identify high-risk hot spots for prioritized inspection and maintenance, given limited resources.
Publisher
ProQuest Dissertations & Theses
Subject
ISBN
9798304912723
