Railway Maintenance Research Articles

Integration of on-board monitoring data into infrastructure management for effective decision-making in railway maintenance

Abstract Railway infrastructure is a crucial asset for the mobility of people and goods. The increased traffic frequency imposes higher loads and speeds, leading to accelerated infrastructure degradation. Asset managers require timely information regarding the current (diagnosis) and future (prognosis) condition of their assets to make informed decisions on maintenance and renewal actions. In recent years, in-service vehicles equipped with on-board monitoring (OBM) measuring devices, such as accelerometers, have been introduced on railroad networks, traversing the network almost daily. This article explores the application of state-of-the-art OBM-based track quality indicators for railway infrastructure condition assessment and prediction, primarily under the prism of track geometry quality. The results highlight the similarities and advantages of applying track quality indicators generated from OBM measurements (high frequency and relatively lower accuracy data) compared to those generated from higher precision, yet temporally sparser, data collected by traditional track recording vehicles (TRVs) for infrastructure management purposes. The findings demonstrate the performance of the two approaches, further revealing the value of OBM information for monitoring the track status degradation process. This work makes a case for the advantageous use of OBM data for railway infrastructure management, and attempts to aid understanding in the application of OBM techniques for engineers and operators.

Robust Lifetime Estimation with Small Sample Sized High Speed Railway ETCS Component Data

Railway maintenance and efficient operation has been an important issue for a safe rail traffic. When an unexpected malfunction occurs on various components on the train or on the railway system, it may result in unscheduled maintenance which may cause the rail traffic to stop. In this paper, we study the random failure model of some frequently malfunctioning high-speed railway equipment based on the statistical analysis of the real data of failure records in Turkey. Popular distribution functions and parameter estimation methods have been used considering that the data has a small sample size, and it may contain outliers. In this study, we showed that for the case of a few numbers of failure data, the L-moments method gives effective results when there exists no outlier and the robust-M method gives effective results when there exists an outlier or outliers.

Three-stage location-routing programming for heavy-duty railway maintenance machinery base based on multi-factor combined weights

Abstract An imperative shift towards a scientific and methodical approach has become increasingly essential when deliberating the placement of maintenance bases for robust railway machinery dedicated to heavy-duty tasks. This article addresses the need for a systematic approach to placing maintenance bases for heavy-duty railway machinery. It introduces a meticulously crafted three-stage location-routing paradigm that factors in both qualitative and quantitative determinants. This approach optimizes routing strategies for the machinery while constructing an evaluative framework for base placement. An evaluative index is created by merging relevant qualitative and quantitative elements, using an enhanced game theory methodology combined with the evidential analytic hierarchy process and the entropy weight method to balance subjective and objective considerations. The model’s complexity is handled with an improved genetic algorithm and a hybrid ant colony algorithm, effectively addressing location and routing interactions. Practical implementation and case studies validate the model’s and algorithm’s efficacy, confirming their applicability in real-world scenarios.

The solar radiation induced spatiotemporal temperature distribution and time-varying temperature effect in long-span ballastless track continuous beam-arch bridge

The influence of a time-varying temperature environment on the mechanical performance and geometric alignment of long-span ballastless track continuous beam-arch bridge is significant. However, the structural temperature loads are often simplified into a single mode, making it difficult to accurately simulate the evolution of geometric alignment of long-span ballastless track continuous beam-arch bridge under time-varying temperature effects. To explore the spatiotemporal distribution characteristics of temperature fields and the evolution patterns of track geometric alignment in long-span ballastless track continuous beam-arch bridge, a refined analysis model for the time-varying temperature field is established incorporating ray tracing methods and real-time shadow techniques. Combining the temperature experiment results, the spatiotemporal distribution characteristic and evolution patterns of temperature across the bridge system and key sections are explored. Building on this foundation, the evolution of track geometric alignment on the bridge is analyzed, and the measures for controlling track geometric alignment are proposed. The research findings indicate that the surface temperature of the structure undergoes periodic changes throughout the day, and the orientation of the bridge significantly influences the temperature field of the beam-track structure. The vertical displacement extremes of the ballastless track on the bridge are positively correlated with the structural temperature. The track maintenance moments affect the track alignment on the bridge, but the extreme deformation of the ballastless track and the temporal variation pattern of the structural system temperature remain consistent. To minimize irregularities in the ballastless track geometric alignment on the bridge, adjustments to the track position during high-speed railway maintenance window should choose the moments close to the daily average temperature, aiming to decrease the effects of time-varying temperature on track irregularity.

Research on ride comfort evaluation method of EMU based on CART algorithm

In this paper, the research team presents the findings of long-term tracking tests on power-centralized EMU. The results demonstrate clear discrepancies between the established ride comfort evaluation results and the actual vibration state of the vehicle body and the subjective experiences of the drivers and passengers. A fuzzy evaluation was employed to establish the data set, while the entropy weight method was used to determine the key indexes. The CART decision tree algorithm was then used to establish a model for the evaluation of ride comfort in power-centralized EMU. The resulting evaluation was verified through dynamic simulation and field testing. The findings of the research demonstrate that: The lateral ride index can be employed as the primary index for evaluating the vibration state of the vehicle. By implementing three consecutive lateral ride comfort evaluations of the vehicle, the false positive rate and false negative rate can be maintained below 4%. The accuracy of the evaluation results is greater than 95% when a CART decision tree model is constructed using the minimum, maximum, standard deviation, and 50th percentile of the lateral ride index of the vehicle. The results of the simulation and field tests demonstrate that the ride comfort evaluation results are highly consistent with the actual vibration state of the moving EMU, the annoyance rate, and the human subjective feeling. This paper presents a comprehensive analysis of the research content, taking into account the subjective experiences of drivers and passengers, as well as the practical considerations of railway maintenance personnel. The findings offer a valuable contribution to the decision-making process of railway maintenance professionals.

Rail Flaw B-Scan Image Analysis Using a Hierarchical Classification Model

AbstractAs railway traffic volumes and train speeds increase, rail maintenance is becoming more crucial to prevent catastrophic failures. This study aimed to develop an artificial intelligence (AI)-based solution for automatic rail flaw detection using ultrasound sensors to overcome the limitations of traditional inspection methods. Ultrasound sensors are well-suited for identifying structural abnormalities in rails. However, conventional inspection techniques like rail-walking are time-consuming and rely on human expertise, risking detection errors. To address this, a hierarchical classification model was proposed integrating ultrasound B-scan images and machine learning. It involved a two-stage approach—model A for fuzzy classification followed by Model EfficientNet-B7 was identified as the most effective architecture for both models through network comparisons. Experimental results demonstrated the model's ability to accurately detect rail flaws, achieving 88.56% accuracy. It could analyze a single ultrasound image sheet within 0.45 s. An AI-based solution using ultrasound sensors and hierarchical classification shows promise for automated, rapid, and reliable rail flaw detection to support safer railway infrastructure inspection and maintenance activities.

ANÁLISE DE ESTRATÉGIAS PARA DISPONIBILIDADE DE MATERIAIS EM MANUTENÇÃO DE MÁQUINAS DE VIA: UMA REVISÃO SISTEMÁTICA

This systematic research addresses the critical importance of material availability for the effective maintenance of track machines, a fundamental element to ensure safety and efficiency in railway operations. The central problem investigated focuses on the challenges faced in provisioning these materials, including logistical issues, stock errors, and delivery delays, which can compromise timely and efficient maintenance. The aim of this study is to analyze innovative and effective strategies that can optimize the provisioning of materials necessary for track machine maintenance. To achieve this goal, a bibliographic research methodology was adopted, focusing on the qualitative analysis of 19 literatures published between 2018 and 2023. The results reveal a convergence among authors on the need to integrate strategic planning, advanced technologies such as artificial intelligence, and efficient inventory management to overcome provisioning challenges. Prescriptive maintenance strategies and the use of advanced diagnostic technologies are highlighted as key elements for more proactive and less reactive maintenance. The final considerations demonstrate that the research problem has been thoroughly explored, with the proposed objectives being achieved. For future research, evaluating the real impact of emerging technologies on the life cycle of railway maintenance is suggested, aiming to provide effective strategies for the continuous optimization of maintenance operations.

Observing and identifying fouled ballast bed using infrared thermography (IRT): A real-time temperature prediction study based on an enhanced BiGRU model

The ballast bed constitutes the cornerstone of the ballasted track. A fouled ballast bed poses a significant threat to its performance, potentially resulting in severe consequences. In recent years, studies have shown that infrared thermography (IRT) technology has emerged as a promising method for detecting the fouled ballast bed. The surface temperatures of clean and fouled ballast beds differ because of their distinct thermodynamic properties. To effectively utilize temperature for identifying the fouled ballast bed, it is essential to accurately predict surface temperatures for two threshold levels of fouling in real-time. To address the issue, this paper proposes an improved BiGRU model (CBGA), and the main contributions are as follows. First, a formula for the surface heat flux density of the ballast bed was derived. In conjunction with existing research findings, the key factors affecting its surface temperature were identified as inputs to the neural network, including the solar radiation intensity, air temperature, wind speed, and humidity. Next, thermodynamic finite element models were established based on a field experiment, which can be utilized to expand the sample library. Leveraging this groundwork, 430 days of temperature data and meteorological data were acquired to train the neural network. Before inputting data into the BiGRU model, CNN and Attention mechanisms were employed to extract local and significant features. Furthermore, a residual network was introduced to ensure the model’s performance. It exhibits superior performance compared to other models. Subsequently, the CBGA model was used to study the impact of different time steps on the prediction accuracy. It was found that a time step of 660 minutes resulted in the best predictive performance. At this time, the evaluation indicators on the testing dataset were: MSE = 0.057, RMSE = 0.238, MAE = 0.168, and MAPE = 0.008. Finally, the reliability and feasibility of the CBGA model were validated using experimental data. These findings demonstrate that the proposed method can achieve real-time prediction of the ballast surface temperature, laying a solid foundation for the practical application of IRT technology in railway maintenance.

Experimental Analysis of Dynamic Vibrations in Rails by Using Different Types of Vibration Absorption Pads

Railway pads are a type of component that is used in railway construction and maintenance. They are typically made of rubber or similar material and are placed between the railway sleeper and the rail itself. Railway pads provide a layer of cushioning and insulation between the rail and the sleeper, which helps to reduce noise and vibration from passing trains. Railway pads can also help extend the rail's life by reducing the wear and tear caused by constant friction between the rail and sleeper. They are especially useful in areas where noise pollution is a concern, such as residential areas or near hospitals and schools. Several types of railway pads are available, including continuous, discrete, and ribbed designs. The specific type of pad used will depend on factors such as the type of rail being used, the expected loads and speeds of passing trains, and the requirements of the railway system. Overall, railway pads are an important component in the construction and maintenance of railway systems, helping to reduce noise and vibration while also prolonging the life of the rail and other components. In this paper, we are testing three types of vibration rubber pads i.e., Ethylene Propylene Diene Monomer (EPDM), High-Density Polyethylene (HDPE), nitrile and comparing three vibration frequencies. The dynamic frequency of welded rails with HDPE pads has having 1190 Hz mean value. The dynamic frequency of rail tracks with single- and double-layer EPDM pads their mean values are 499 Hz and 335 Hz respectively. The experimental values clearly show that single- and double-layer EPDM pad dynamic frequencies are reduced by up to 58% and 71%, respectively, over the HDPE pad.

Vulnerability Evolution Mechanism of High-Speed Railway Construction System in the Goaf Site of Qinshui Coalfield

The main reason for serious consequences caused by disasters is “insufficient emergency preparedness and obvious system vulnerability”. To ensure the safe operation and maintenance of the high-speed railway in goaf sites and achieve effective hierarchical management, the vulnerability theory was introduced, based on the system approach, to explore the characteristics of hierarchical evolution. Firstly, based on the analysis of the internal and external disturbances of the high-speed railway construction system with underlying goaf areas, the vulnerability characteristic elements were defined, and a conceptual vulnerability model was proposed. Secondly, based on the literature research, accident cases, and standards and specifications, 25 factors affecting vulnerability were identified and a two-dimensional cross matrix was used to establish a vulnerability evaluation index system. Then, ISM was used to build hierarchical relationships among indicators, analyze the vulnerability characteristics of direct factors, indirect factors and essential factors, and a vulnerability evolution analysis framework was proposed. Finally, the Taiyuan–Jiaozuo high-speed railway project was selected for hierarchical and progressive vulnerability management. The study analyzed the coupling effect and evolution mechanism of various elements of the safety management system under external disturbances, identified the weak links of the high-speed railway system in the goaf site, and provided a decision-making basis for continuous dynamic optimization and emergency response technology, which complied with the strategic policy of sustainable development.

An Automated Rack Storage System for Railway Devices and Equipment Using QR Code and Cloud Database Technology

This research project aims to develop an automated storage rack system for railway devices and equipment using QR codes and cloud-based database technology. The automated rack system can precisely store railway devices and equipment using the cloud-based QRC/QR Code database. It explores the drawbacks of conventional storage management methods in the railway industry, like lack of order, inefficiency, and security issues. Using QR codes to track and monitor all tools, equipment, and components, the proposed system includes a cloud-based database providing real-time inventory visibility and data analysis capabilities. Additionally, the system features CCTV surveillance for added security level and mobile phone integration to ensure that only you, as the end user, get access. Based on the study's findings, automated storage rack systems promise significant improvements in efficiency, security, and operational effectiveness for railway maintenance.

Enhancing railway track maintenance with real-time ultrasonic and moisture sensing: Proactive and zone-specific management strategies

The incorporation of ultrasonic and moisture sensors represents a significant advancement in railway track maintenance, offering a comprehensive and efficient means of assessing conditions, reducing time, and eliminating manual labor. Mounted on a trolley, these sensors simultaneously measure ballast cushion availability, moisture content, and humidity, providing real-time data stored in a cloud-based account for effective management. This data enables proactive maintenance, calculating ballast requirements using artificial intelligence during normal trolley runs. The system enhances safety and performance by assessing ballast availability and deficiencies along the track. Variations in defect lengths are evident between zones; for instance, the mud pumping zone ranges from 2 to 38 m with an average of 19 m and peak moisture content at 35 %. In contrast, zones with high freightage and track density exhibit more consistent defect lengths, averaging 13 and 10 m respectively, with moisture contents between 5 % and 15 %. The study emphasizes the necessity for tailored maintenance plans for each zone to ensure railway infrastructure effectiveness and safety.

A novel linetype recognition algorithm based on track geometry detection data

Abstract The detection and recognition of track geometry are crucial for ensuring railway transportation safety and maintenance efficiency. Traditional linetype recognition algorithms struggle to effectively handle the complexities of variable track geometry and noise interference. To address these challenges, a novel linetype recognition algorithm (NLRA) is proposed, which thoroughly analyzes the characteristics of track curvature across straight lines, transition curves, and circular segments. By employing partial differential equations for data preprocessing, NLRA effectively filters out noise. Utilizing curve design parameters and integrating line anti-noise recognition with multi-segment line fitting techniques, the algorithm accurately identifies each line segment and incorporates a single-point recognition method for data gaps. Experimental results demonstrate that NLRA outperforms traditional algorithms in terms of real-time performance, recognition accuracy, anti-interference capability, and noise handling. Notably, NLRA enhances accuracy by at least 35.1% and recall by at least 17.4% on selected typical routes. This work underscores the algorithm's potential to provide efficient and accurate tools for track maintenance, facilitate advancements in track detection software, and significantly contribute to railway transportation safety.

Framework of Scan to Building Information Modeling for Geometric Defect Localization in Railway Track Maintenance

Railway transportation plays a vital role in modern society, enabling the safe and efficient movement of people and goods over long distances. To ensure the longevity and safety of a railway infrastructure, the regular maintenance of tracks is crucial. Traditional track inspections, conducted manually to monitor geometric parameters and to identify defects, are time-consuming, labor-intensive, and prone to human error. Current Scan-to-BIM frameworks for railway maintenance also lack standardized methods for extracting geometric parameters that can be easily integrated into Building Information Modeling (BIM). Additionally, the Industry Foundation Classes (IFC) standard, used for BIM data exchange, does not support storing parameter values at specific chainage points along the track, limiting defect localization. A framework is proposed to address these challenges by standardizing the extraction of geometric parameters from point cloud data and ensuring seamless integration with BIM. The framework calculates parameters at station chainage points and generates additional chainage points along the track, associating the data with the corresponding chainage. A case study demonstrates the framework’s ability to enhance defect localization, using the EN 13848-5 European Standard to identify defects at specific chainages. Ultimately, this approach contributes to the more effective lifecycle management of railway tracks.

Research on a Joint Extraction Method of Track Circuit Entities and Relations Integrating Global Pointer and Tensor Learning.

To address the issue of efficiently reusing the massive amount of unstructured knowledge generated during the handling of track circuit equipment faults and to automate the construction of knowledge graphs in the railway maintenance domain, it is crucial to leverage knowledge extraction techniques to efficiently extract relational triplets from fault maintenance text data. Given the current lag in joint extraction technology within the railway domain and the inefficiency in resource utilization, this paper proposes a joint extraction model for track circuit entities and relations, integrating Global Pointer and tensor learning. Taking into account the associative characteristics of semantic relations, the nesting of domain-specific terms in the railway sector, and semantic diversity, this research views the relation extraction task as a tensor learning process and the entity recognition task as a span-based Global Pointer search process. First, a multi-layer dilate gated convolutional neural network with residual connections is used to extract key features and fuse the weighted information from the 12 different semantic layers of the RoBERTa-wwm-ext model, fully exploiting the performance of each encoding layer. Next, the Tucker decomposition method is utilized to capture the semantic correlations between relations, and an Efficient Global Pointer is employed to globally predict the start and end positions of subject and object entities, incorporating relative position information through rotary position embedding (RoPE). Finally, comparative experiments with existing mainstream joint extraction models were conducted, and the proposed model's excellent performance was validated on the English public datasets NYT and WebNLG, the Chinese public dataset DuIE, and a private track circuit dataset. The F1 scores on the NYT, WebNLG, and DuIE public datasets reached 92.1%, 92.7%, and 78.2%, respectively.

Optimizing Railway Tribology: A Systematic Review and Predictive Modeling of Twin-Disc Testing Parameters

Twin-disc testing is crucial for understanding wheel–rail interactions in railway systems, but the vast array of testing parameters and conditions makes data interpretation challenging. This review presents a comprehensive analysis of the twin-disc literature experimental data, focusing on how various parameters influence friction and wear characteristics under stationary contaminant conditions. We systematically collected and analyzed data from numerous studies, considering factors such as contact pressure, speed, material hardness, sliding speeds, adhesion, and a range of contaminants. This research showed inconsistent data reporting across different studies and statistical analyses revealed significant correlations between testing parameters and wear rates. For sand-contaminated tests, a correlation between particle size and flow rate was also highlighted. Based on these findings, we developed a simple predictive model for forecasting wear rates under varying conditions. This model achieved an adjusted R2 of 0.650, demonstrating its potential for optimizing railway component design and maintenance strategies. Our study provides a valuable resource for researchers and practitioners in railway engineering, offering insights into the complex tribological interactions in wheel–rail systems and a tool for predicting wear behavior.

Numerical parametric study for ballast assessment using SVM applied to GPR data

Abstract This study explores novel radar-based methodologies for ballast fouling evaluation and structural anomalies in railway structures. Ground Penetrating Radar (GPR), coupled with numerical modeling, is utilized to provide insights into ballast condition and structure. Various data processing methods, including the Matrix Pencil Method (MPM) and Full Waveform Inversion (FWI), are investigated for their effectiveness in detecting fouling. Support Vector Machine (SVM)-based machine learning approaches are employed to enhance classification accuracy. The results demonstrate that integrating raw GPR signals with MPM yields the most accurate classification results, facilitating efficient assessment of ballast condition and contributing to improved railway maintenance strategies.

A model-driven approach for prospective ergonomics: application to ikigai robotics

Prospective Ergonomics requires building a vision of the future, which can be achieved empirically (e.g. analysing unmet needs) and/or creatively (e.g. creating future needs). We develop an alternative way of imagining the future, through a model-driven approach. Based on several developmental models, we provide a global picture of possible future(s) emphasising higher-ordered motivations and values (e.g. meaningfulness, accomplishment). To implement them, we then present a model of human accomplishment reinterpreting the concept of ikigai in light of selected psychological theories (e.g. self-determination, eudaemonic well-being, mindfulness). Finally, we apply it to an Industry 5.0 case study named ikigai robotics: we designed an equipment for railway maintenance following a double design process – a functional design loop and a motivational design loop. The process proved inspirational and the results both original and promising, opening avenues for Prospective Ergonomics to develop a new approach for designing the future.

Intelligent prediction of rail corrugation evolution trend based on self-attention bidirectional TCN and GRU

Analyzing the evolution trend of rail corrugation using signal processing and deep learning is critical for railway safety, as current traditional methods struggle to capture the complex evolution of corrugation. This present study addresses the challenge of accurately capturing this trend, which relies significantly on expert judgment, by proposing an intelligent prediction method based on self-attention (SA), a bidirectional temporal convolutional network (TCN), and a bidirectional gated recurrent unit (GRU). First, multidomain feature extraction and adaptive feature screening were used to obtain the optimal feature set. These features were then combined with principal component analysis (PCA) and the Mahalanobis distance (MD) method to construct a comprehensive health indicator (CHI) that reflects the evolution of rail corrugation. A bidirectional fusion model architecture was employed to capture the temporal correlations between forward and backward information during corrugation evolution, with SA embedded in the model to enhance the focus on key information. The outcome was a rail corrugation trend prediction network that combined a bidirectional TCN, bidirectional GRU, and SA. Subsequently, a multi-strategy improved crested porcupine optimizer (CPO) algorithm was constructed to automatically obtain the optimal network hyperparameters. The proposed method was validated with on-site rail corrugation data, demonstrating superior predictive performance compared to other advanced methods. In summary, the proposed method can accurately predict the evolution trend of rail corrugation, offering a valuable tool for on-site railway maintenance.

Knowledge Graph Completion for High-Speed Railway Turnout Switch Machine Maintenance Based on the Multi-Level KBGC Model

The incompleteness of the existing knowledge graphs in the railway domain creates information gaps, impacting their quality and effectiveness in the operation and maintenance of high-speed railway turnout switch machines. To address this, we propose a multi-layer model (KBGC) that combines KG-BERT, graph attention network (GAT), and Convolutional Embedding Network (ConvE) for knowledge graph completion in railway maintenance. KG-BERT fine-tunes a pre-trained BERT model to extract deep semantic features from entities and relationships, converting them into graph structures. GAT captures key structural relationships between nodes using an attention mechanism, producing enriched semantic and structural embeddings. Finally, ConvE reshapes and convolves these embeddings to learn complex entity interactions, enabling accurate link prediction. Extensive experiments on the HRTOM dataset, containing triplet data from high-speed railway turnout switch machines, demonstrate the model’s effectiveness, achieving an MRR of 50.8% and a Hits@10 of 60.7%. These findings show that the KBGC model significantly improves knowledge graph completion, aiding railway maintenance personnel in decision making and preventive maintenance, and providing new tools for railway maintenance applications.