Railway track health monitoring and maintenance are crucial stages in railway asset management, aiming to enhance the train operation quality and service life. For this aim, various inspection means (using diverse non-destructive testing techniques) have been applied, however, these means are mostly not able to monitor whole railway track network or track underlying layers (e.g., ballast and subgrade). The use of remote sensing techniques, such as Interferometric Synthetic Aperture Radar (InSAR), can expedite the defect diagnosis process for railway tracks, elevating the scope of health monitoring to a network-wide level. The Ground Penetrating Radar (GPR) has emerged as a particularly reliable method, especially for detecting structural deficiencies in underlying layers. As a result, combining the two distinct non-destructive testing techniques – GPR and InSAR – presents a promising strategy for efficient railway asset management. Recognizing the significance of embracing newer and more advanced monitoring strategies, this paper reviews the fusion of GPR and InSAR methodologies, and explores the potential integration of machine learning models to develop a predictive health monitoring and condition-based maintenance approach for railway tracks.

This study explores the use of infrared thermography (IRT) technology for the non-destructive evaluation of ballast fouling in railway tracks, focusing on the influence of parent rock types and fouling materials. Utilizing thermal imaging, the research investigates how variations in ballast conditions affect surface temperature, which serves as an indicator of structural integrity and health. The experimental setup involved ballast samples derived from three different rock types—basalt, limestone, and andesite—fouled with commonly encountered materials like sand and clay at varying percentages. Results demonstrate that fouling level and type significantly influence the thermal signatures captured by IRT passive camera. Notably, ballast derived from darker rocks exhibited higher temperatures, indicating greater emissivity, while fouled ballast showed distinct temperature patterns compared to clean samples, emphasizing the potential of thermal imaging in detecting and quantifying fouling in ballast layers. This research underscores the viability of IRT passive camera in the routine maintenance and monitoring of railway infrastructure, providing a foundation for further development of integrated diagnostic tools for railway management systems.

Parent rock strength and crumb rubber modification are two critical mechanical parameters that significantly decide the ballast layer degradation subjected to train dynamic loading. Using machine learning to predict ballast degradation considering these two parameters is helpful for deciding ballasted track maintenance cycle. In the current study, the ballast degradation process data (variables: parent rock types, loading types, ballast gradations and compositions of crumb rubber-ballast mixture) were used to train machine learning models. The drop-weight impact loading tests were performed to simulate different train dynamic loadings. Two well-established machine learning models, i.e., random forest (RF) and support vector regression (SVR) were trained and verified, to more effectively assess the importance of these variables. The results from the validated machine learning models confirm that the parent rock type is the most influential parameter, followed by the loading type (applied stress level), to control and predict the degradation of the ballast-CR mixture. The experimental assessment reveals that although the incorporation of CR suppresses degradation across all characterized rock types, the improvement in performance of the ballast-CR specimen against degradation is more noticeable for high-strength parent rock subjected to a considerable stress level. Meanwhile, this positive influence is also observed for ballast of weaker strength when the applied stress level is low.

The occurrence of ballast contamination or fouling frequently results in a sudden decline in the capacity of railway ballasted tracks. Considering the various sources of ballast fouling, clay is the most severe one for causing a drastic reduction in the drainage capacity of the ballast layer. In the current study, we utilized a large-scale flume test to measure the water height along the cross-section of the clay-fouled ballast. Subsequently, an analytical–numerical (A-N) approach was developed to simulate the movement of water through porous media under steady-state conditions, while also considering the flow regime. This A-N approach was validated using the results of flume tests. Finally, the validated A-N approach was employed to generate a dataset and develop machine learning models for predicting water height. The characterized machine learning models included random forest regression (RFR), support vector machine (SVM), and extreme gradient boosting (XGBoost). Various variables, such as ballast gradation, fouling ratio, bed slope, rainfall rate, and water height on the side ditch, were incorporated into the machine learning models to reveal the contribution of each individual variable. Results show that for clean ballast, the incorporation of a nonlinear model between flow velocity and hydraulic gradient in the A-N approach is crucial to properly estimate the experimental measurements. However, a comparison of the water height measured via the flume test and the water level estimated based on the A-N approach confirms the suitability of the linear model, i.e., Darcy's law, for the water flow regime through clay-fouled ballast. According to the machine learning results, particularly those from the XGBoost model, which was characterized as the elite model, the rainfall rate and the fouling index emerged as the most influential variables affecting the water height in the clay-fouled ballast layer of the railway track.