Railway track quality, safety and reliability through train-track dynamics and advanced machine learning - Clarke Lecture 2023

Aug 3, 2026

Upload by Patrick

Railway tracks experience dynamic loads and environmental influences that directly affect track quality, operational safety, and overall system reliability. The work investigates how track characteristics and quality impact performance under real operational conditions, and identifies key factors driving degradation and reliability risks. Advanced machine learning is leveraged to model these dynamics, enabling real-time assessment of track conditions and improving ride-comfort evaluation. Results show strong performance from CNN variants, with dilated CNN achieving the best predictive fit and supporting robust, efficient computation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages2

Reading Time5 min

AccessFree to download

Research Questions

Contextual Background

Methods and Results

The Findings in Context

Conclusions

Key Publications

University of Birmingham Railway track quality, safety and reliability through train-track dynamics and advanced machine learning Huang, Junhui; Kaewunruen, Sakdirat License: Creative Commons: Attribution (CC BY) Citation for published version (Harvard): Huang, J & Kaewunruen, S 2023, 'Railway track quality, safety and reliability through train-track dynamics and advanced machine learning', Clarke Lecture 2023, Birmingham, United Kingdom, 22/06/23-22/06/23 . Link to publication on Research at Birmingham portal General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document. When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, [please contact UBIRA@lists.bham.ac.uk](please contact UBIRA@lists.bham.ac.uk) providing details and we will remove access to the work immediately and investigate. Download date: 02. Aug. 2026 Railway track quality, safety & reliability through traintrack dynamics and advanced machine intelligence Junhui Huang, Dr Sakdirat Kaewunruen, School of Engineering Research Questions 1. How do characteristics, quality and reliability of railway tracks impact system performance? 2. What are the key factors influencing the tracks in dynamic operational conditions? 3. What benefits of leveraging AI in enhancing the performance of track? 4. How can AI be applied to determine real-time track conditions? Contextual Background Railway tracks are subjected to dynamic forces and environmental factors. AI can be used to monitor traintrack Methods Results 1. Both the performance of the standard CNN and dilated CNN are above 0 .90 . 2. Considered factors: two different model types, three sample size, irregularities included. 3. Dilated CNN produces the best R2 of 0 .97 on the longest sample size. The Findings in Context New mobile-phone App for Real-time Ride Comfort Assessment AI models provide satisfied performance Fast calculation with full technical robustness Easy to implement and to connect real-time via 5G Reducing cost, and Empowering passengers New method Robust Easy to use Accurate Conclusions 1. We have proposed a dilated CNN using the sensory data acquired from axle box which is immune to speed up to 120 km/h and bad weather conditions as there are sophisticated accelerometers designed to tackle the extreme weather conditions. 2. We have confirmed that the performance of the proposed model is highly satisfied, the dilated technique contributes to saving computational cost, and the consideration of irregularities makes no significant difference to the model's performance. 3. An implication of this study is the possibility that wireless accelerometers can be mounted onboard service trains. 4. New smartphone App has been developed to assess real-time ride comfort using machine intelligence. Key

FAQs

How do railway track characteristics affect system performance and reliability?

The research addresses how track quality and reliability influence system performance, emphasizing the role of track-related characteristics in overall operational outcomes under dynamic conditions.

What AI approach is proposed for determining real-time track conditions?

A dilated CNN model is proposed using sensory data acquired from the axle box, designed to remain effective under speed up to 120 km/h and adverse weather conditions.

What performance results and practical outputs does the study provide?

The model yields strong accuracy (reported above 0.90 for CNN variants, with dilated CNN achieving best R² up to 0.97), and the study supports a mobile-phone app concept for real-time ride comfort assessment with fast, robust computation.

Aug 3, 2026

Upload by Patrick