Infrastructure management is one of the main problems faced by railway managers. Generally, decision making is based on track geometry measurements made with recording machines. This perspective is able to detect where the problems are and their magnitude, but does not provide information on the origin of the defects. To alleviate this diagnostic shortcoming, approaches such as multi-domain diagnosis, which takes into account multiple parameters, have been developed. However, the developers of this technique point out that it is necessary to perform an individualized analysis so that the programming of interventions is as effective as possible and lasts over time. It is on the basis of this problem that the need arises to create information that complements what already exists, which is why it has been decided to develop performance indicators that provide more information than a traditional parameter or indicator. In this research work, the functional and reliability method has been used to obtain the performance of granular components, together with expert analysis and multi-criteria data analysis. To assess the methodology, a 35 km long single-track study site was evaluated. This road has all the input parameters for the implementation of the developed methodology and, in addition, the multi-domain diagnostic analysis was taken into account as an operational reference classification. The results indicate that the indicators developed are sensitive and appropriately matched to decision making; that is, when an indicator is categorized as good performance, no intervention on the road is necessary, whereas, poor performance states are predominantly associated with intervention cases. Furthermore, this individualized approach supports the identification of specific layers likely linked to observed defects, facilitating a better interpretation of the underlying degradation mechanisms within the railway substructure. By supporting the identification of affected layers, the methodology informs targeted corrective actions and fosters more resilient maintenance planning. While its practical application relies on integrating complementary inspection data and multi-diagnostic techniques, the proposed framework is designed specifically as a specialized module for substructure diagnosis. Consequently, it should be viewed as a high-level decision-support tool for infrastructure management rather than a standalone system for the entire track assembly.

This paper presents a framework for simulating minimum-time overtaking manoeuvres between two race cars on a three-dimensional (3D) race track. The overtaking problem is formulated as a dynamic game while avoiding collisions. To accurately capture the influence of track geometry on vehicle behaviour, recent developments in track-surface modelling are incorporated. The simulations are performed on the Darlington Raceway, which is characterised by significant camber angles and large lateral camber angle variations. A realistic overtaking scenario comprises identical cars, with the lead car fitted with worn tyres that degrade its performance. Two key challenges are addressed: the formulation of a geometry-based inequality constraint to prevent inter-vehicular collisions, and the implementation of interlaced receding-horizon optimal control problems that constitute a Stackelberg game. The game comprises continuous moves that are played simultaneously and discretely.

In urban rail transit systems, passenger ride comfort (PRC) significantly influences ridership levels. Various factors affect PRC, with track geometry playing a crucial role. Among geometry parameters, the minimum required tangent length between consecutive railway curves is particularly significant. While this parameter affects passenger comfort considerably, it has received less attention in current design practices. This study addresses this gap by optimizing tangent length requirements to improve passenger comfort. To this end, a vehicle-track interaction model was established and subsequently validated through comprehensive field measurements. The influence of curve radius, tangent length, and speed on the PRC was investigated through a parametric analysis, considering various curve configurations. As a result, a large data bank was created, and a model was developed using data mining techniques to predict minimum tangent lengths. The accuracy and computational performance of the model were discussed. The results obtained indicate that reverse curves require 5% longer tangent lengths and have a 9% higher ride comfort index compared to those of compound curves. The model optimizes tangent lengths for speeds above 80km/h, leading to improved ride comfort. This improvement contributes to increased rail transit ridership.

High-speed railway (HSR) systems operating at velocities up to 350 km/h impose stringent requirements on vehicle–track interaction, wheel–rail contact mechanics, and infrastructure maintenance strategies. At such speeds, railway systems behave as highly coupled nonlinear systems. Small variations in contact conditions or track geometry can significantly amplify dynamic responses. This paper presents a comprehensive review of the theoretical foundations of wheel–rail rolling and sliding contact, with particular emphasis on wear and rolling contact fatigue (RCF). Global research trends in railway dynamics are reviewed, highlighting the increasing integration of multibody dynamics (MBD), advanced contact models, and damage prediction techniques. The applicability of these methodologies to the future high-speed railway system in Vietnam is discussed, and engineering-oriented recommendations and parametric considerations are proposed for the development of simulation-based design and predictive maintenance frameworks.

Derailment, one of the most critical accidents in railway systems, is directly associated with running safety. The derailment coefficient, which is widely used as a representative quantitative indicator for derailment evaluation, is determined by the interaction between the vehicle and track and is closely related to track conditions. Previous studies have investigated the effects of operational conditions, such as curve radius, cant, and speed, as well as track conditions, on the derailment coefficient and running safety. However, most of these studies are limited to analytical approaches involving specific conditions or analyses based on limited field-measurement data. This study proposes an artificial intelligence (AI)-based model to predict the derailment coefficient using field-track inspection data. The input variables comprise track geometry parameters, including the alignment, gauge, cross-level, twist, and longitudinal level, all obtained from track inspection vehicles, which effectively reflect actual track conditions. The results demonstrated that the proposed AI-based model effectively captures the complex nonlinear relationships between track inspection data and the derailment coefficient, thereby achieving high prediction accuracy.

Railway Track Geometry Irregularity Exceedance Prediction Based on CNN-BiLSTM-Attention and Neural-Wiener Process Fusion with Degradation Feature Diversity

During the study, a nickel–chromium-based coating was formed on structural steel S235 using flame spraying technology. A NiCrMoSiBFeCuC powder mixture with the following composition was used for coating formation: ~67% Ni; 0.45% C; 4.11% Si; 3.92% B; 2.72% Fe; 15.7% Cr; 1.94% Cu; 4.69% Mo. The sprayed coatings were remelted using a gas torch at a temperature of 1100 °C in air. At the next stage of coating formation, the samples were processed with a fiber laser under four processing speed modes - without beam oscillation and with oscillation amplitudes of 1 mm and 3 mm. The aim of this study was to investigate the effect of laser processing on NiCrMoSiBFeCuC coatings formed by spraying followed by gas torch remelting. The shape and geometry of the laser-remelted tracks, microhardness of the samples, as well as tribological properties and wear resistance were investigated. The combined method of coating formation, including spraying followed by laser treatment, allows the formation of a continuous or localized NiCrMoSiBFeCuC coating layer with increased microhardness and wear resistance on the surface of the component

Non-disruptive rail track geometry measurement system using an unmanned aerial vehicle and a light detection and ranging sensor

Railway track geometry degradation: A review of prediction methods

Abstract Infrastructure managers rely on diagnostic trains that periodically measure track geometry and vehicle accelerations to ensure the safety of the railway network. Their runs are scheduled depending on the line priority, in order to safely monitor the evolution of track defects. However, sudden and unpredictable defect growth may happen and be missed between successive runs. Therefore, condition monitoring systems have been installed on in-service vehicles. In fact, these trains run every day along the same line, so they can provide additional information useful for maintenance practices. When trains run along conventional lines, their speed significantly changes depending on the line characteristics, and vehicle accelerations strongly depend on speed. Therefore, monitoring systems that rely on vehicle accelerations should carefully take this effect into account. In this paper, a methodology to estimate the track longitudinal level using bogie accelerations from an in-service vehicle is presented. The recorded accelerations were double-integrated to account for the speed variation, and a model-based strategy was adopted to reduce the filtering action of the primary suspension. Data were recorded during a two-year monitoring campaign along an Italian railway line. The methodology allowed for the estimation of the longitudinal level along specific track sections, considering statistical measures like the peak value. A maximum error of 1 mm was found between the estimated values and those measured by the diagnostic train (considering a defect with magnitude of 7.5 mm). Therefore, the results showed that it is possible to estimate the peak longitudinal level between the two rails using one single vertical accelerometer installed on the bogie of an in-service vehicle. The results of this research may be used to support the current maintenance strategy with daily estimations of track longitudinal level. It should be noted that specific attention was given only to this type of track geometry parameter, since it often drives maintenance operations. In the future, the possibility to extend the methodology to the estimation of different type of defects, like cross-level and twist, could be considered.

This study presents a new method to mitigate the effects of train speed on dynamic measurement data acquired by high-speed diagnostic trains for railway track geometry evaluation. Variations in train velocity are shown to significantly distort signal characteristics, such as wavelength and amplitude, which can hinder accurate defect detection. To overcome this limitation, a speed-independent data processing technique is developed that transforms acceleration signals into displacement in the spatial domain, effectively normalising the data across different speeds. The proposed approach is validated using vertical acceleration data collected from axle box-mounted sensors on a Turkish State Railways (TCDD) high-speed diagnostic train. Experimental results confirm that the method significantly reduces speed-induced discrepancies, enhancing the consistency and reliability of track condition assessments. The study further demonstrates that consistent sensor placement, vehicle orientation, and measurement direction are critical to ensuring data comparability. The findings contribute to more accurate infrastructure monitoring and support improved maintenance planning in high-speed rail networks.

Point cloud technology, characterized by its high-precision 3D geometric acquisition in complex railway environments, has become a cornerstone for the intelligent detection, monitoring, and maintenance of railway infrastructure. This paper provides a systematic review of point cloud applications across critical railway scenarios, encompassing track geometry extraction, infrastructure component identification, tunnel and bridge modeling, clearance and encroachment analysis, and structural condition monitoring. We evaluate various mobile and stationary acquisition platforms alongside their typical data processing workflows. Furthermore, this review synthesizes cutting-edge advancements in processing algorithms, with a focus on feature extraction, semantic segmentation, and the transformative impact of deep learning and artificial intelligence on data fusion. Notably, the paper explores the synergy between point clouds and computational mechanics, specifically the construction of high-fidelity digital twins through multi-physics coupling to enable real-time simulation of structural stress distribution and damage evolution. We critically analyze persistent technical bottlenecks, such as acquisition efficiency, monitoring precision, data fragmentation, environmental interference, and the complexities of multi-modal data fusion. Finally, the paper outlines future research trajectories, focusing on autonomous intelligent sensing, multi-sensor integration, and the comprehensive digital transformation of railway infrastructure management, aiming to provide a robust theoretical framework and technical roadmap for the sustainable intelligentization of global railway systems.

• Transition zones cause track geometry issues due to abrupt stiffness changes. • Self-levelling sleepers restore ballast contact and correct settlement up to 40 mm. • SLS help control hanging sleepers and reduce differential settlement. • Polymeric sleepers improve load distribution and durability in track systems. • Modular SLS offer a low-disruption, cost-effective solution for transition zones. Railway track transition zones present engineering challenges due to their abrupt change in stiffness between structural elements such as embankments, bridges and tunnels affecting track geometry parameters. Although a variety of stiffness-based remedial measures have been widely applied, their implementation can be constrained by high capital cost, operational disruption, and the complexities associated with modifying the substructure. As a result, interventions in practice commonly focus on controlling permanent deformations and differential settlement, particularly related to the development of hanging sleepers. Thus, this study investigates the use of modular self-levelling sleepers (SLS) as a solution. To do so, two concept SLS systems are designed and developed: one employing a granular mechanism (SLS-G), and the other based on a horizontally acting wedge mechanism (SLS-HW). Both variants use the polymeric sleepers and are designed for compatibility with conventional ballasted track systems. Experimental laboratory testing is undertaken, and it is found that the SLS prototypes were able to restore the sleeper-ballast contact for voids up to 40 mm depth, while stress measurements at the interface indicated improved load distribution under the rails. The findings support the proof-of-concept that self-levelling sleepers have the potential to be a modular, low-disruption solution for mitigating track geometry degradation and reducing maintenance requirements at transition zones.

Rail transport is central to achieving sustainable and energy-efficient mobility, and its digitalization is accelerating the adoption of condition-based maintenance (CBM) strategies. However, existing maintenance practices remain largely reactive or rely on limited rule-based diagnostics, which constrain safety, interoperability, and lifecycle optimization. This survey provides a comprehensive and structured review of Artificial Intelligence techniques applied to the preventive, predictive, and prescriptive maintenance of railway infrastructure. We analyze and compare machine learning and deep learning approaches-including neural networks, support vector machines, random forests, genetic algorithms, and end-to-end deep models-applied to parameters such as track geometry, vibration-based monitoring, and imaging-based inspection. The survey highlights the dominant data sources and feature engineering techniques, evaluates the model performance across subsystems, and identifies research gaps related to data quality, cross-network generalization, model robustness, and integration with real-time asset management platforms. We further discuss emerging research directions, including Digital Twins, edge AI, and Cyber-Physical predictive systems, which position AI as an enabler of autonomous infrastructure management. This survey defines the key challenges and opportunities to guide future research and standardization in intelligent railway maintenance ecosystems.

Extreme high-speed laser material deposition (EHLA) is a laser-based coating process known for its high deposition rates and excellent resource efficiency. To further increase productivity, higher laser powers and powder feed rates are increasingly used, but this also raises the thermal load on the processing head due to back-reflected radiation. Inclining the processing head or substrate helps deflect this radiation, while also enabling coatings in geometrically constrained areas, such as internal surfaces. However, inclination introduces asymmetries in the powder and laser distribution, potentially affecting track formation and melt pool behavior. This study investigates how inclination between the processing head and a rotationally symmetric substrate influences track geometry, surface quality, dilution, and thermal impact. Two inclination configurations were examined across a range of angles, with systematic variation of feed and rotation direction to simulate pushing and pulling deposition strategies. The results reveal that inclination has a significant and configuration-dependent effect on surface roughness, build-up volume, powder catchment efficiency, and thermal penetration. Variations in asymmetry orientation relative to the motion direction lead to measurable differences in coating properties. The findings offer new insights into optimizing EHLA process parameters for inclined setups, contributing to more robust and application-specific coating strategies.

Fiber laser surface texturing was applied to AA1050 aluminum to improve friction and wear performance of PTFE coatings. A Taguchi L16 design varied texture geometry (square, diamond, hexagon, circle), scanned area ratio (20% to 80%), and laser power (40 to 100 W) prior to primer plus PTFE topcoat deposition (25 to 35 µm). Dry reciprocating sliding against a 6 mm 100Cr6 ball was conducted at 20 N, 1 Hz, and 50 m, and wear track geometry was measured by non-contact profilometry. The non-textured reference exhibited an average COF of 0.143, whereas the lowest mean COF was achieved with diamond 60% and 40 W (0.095) and the highest with hexagon 60% and 100 W (0.156); hexagon 20% and 60 W matched the reference. ANOVA indicated scanned area ratio as the dominant contributor to COF (39.72%), followed by geometry (35.07%) and power (25.21%). Profilometry confirmed reduced coating penetration for optimized textures: the reference wear track was approximately 1240 µm wide and 82 µm deep, compared with 930 µm and 34 µm for square 80% and 40 W, 997 µm and 39 µm for diamond 60% and 40 W, and 965 µm and 36 µm for hexagon 40% and 40 W.

Shared high-speed rail corridors present a growing challenge for cost allocation as multiple operators use the same infrastructure under open-access or cooperative arrangements. Traditional access charge systems often rely on generalized usage metrics, such as train-kilometers or gross tonnage, which do not explicitly account for the effects of train speed and track geometry on infrastructure wear. As a result, these simplified approaches overlook the true contribution of dynamic forces to rail degradation, leading to potential under- or over-recovery of costs among operators. This motivates the need for a technically grounded framework that links measurable train and track parameters to rail replacement costs for equitable cost sharing. The proposed framework builds on the static rail replacement threshold traditionally used by infrastructure managers and extends it to incorporate the effects of train speed, axle load, and track geometry. It first determines the cumulative tonnage threshold for rail replacement by combining static, dynamic, and lateral loads acting on the rail. This threshold is then used as the basis for estimating the service life of the rail under shared operations, where different train types operate over the same segment. The total rail replacement cost is distributed among the operators according to their proportionate contribution to the cumulative loading during the rail’s service life, resulting in an equitable, load-based access charge for each service. A case study of the Palmdale-Burbank segment of the California High-Speed Rail corridor, currently under construction and expected to be shared with the privately operated Brightline West, demonstrates the application of the proposed framework. The case study used publicly available and assumed operating data of the two train systems. The analysis shows that the heavier, faster train (CAHSR) contributes a greater share of total rail wear and therefore incurs a higher per-trip access charge than the lighter, slower Brightline West train service. The analysis also revealed that neglecting the effects of dynamic and lateral loads would overestimate the rail service life by nearly a factor of two, underscoring the importance of accounting for these forces in both maintenance forecasting and cost allocation. These findings confirm that incorporating speed, axle load, and geometry effects provide a more accurate and equitable basis for allocating maintenance and renewal costs in shared high-speed rail operations.

In high-speed and high-density metro operations, track geometry irregularities can significantly excite vehicle-track coupled vibrations, while mileage localization errors distort irregularity profiles and hinder cross-batch comparisons. To address the challenges posed by GNSS-constrained metro inspections, this study proposes a nominal-mileage alignment framework that integrates turnout detection with segmented mileage correction. The method utilizes an inertial reference to convert irregularity signals from the time domain to the spatial domain, extracting turnout features via short-time Fourier transform. It exclusively relies on left and right vertical acceleration channels, employing dual-channel energy fusion with joint constraints on peak prominence and spacing to achieve robust turnout localization. The detected turnouts serve as anchors for constructing a piecewise linear mapping that corrects the mileage axis while preserving the spatial characteristics of track irregularities. A case study on an 11.8 km in-service metro line demonstrates that the proposed correction reduces P95 by 6-7% and RMSE by approximately 20%, confirming the method’s effectiveness in improving localization consistency and waveform agreement. Dataset 2 further validates the repeatability of turnout detection, using both upward and downward data from another line at different times, supporting the method’s stability and accuracy under the in-service operating conditions examined in this study.

ABSTRACT This study focuses on urban railway train parameters and train speeds up to 120 km/h, which examines how curve alignment parameters affect train dynamic responses and structural vibrations in curved floating slab tracks, combining theoretical analysis with simulation, translating nonlinear physical mechanisms into computable engineering formulas. A coupled train–track dynamic simulation model and an environmental vibration simulation model are established. Key findings show that vertical loads on inner rail fasteners increase with higher unbalanced superelevation, while those on outer rail fasteners decrease. Lateral loads and resultant lateral/vertical loads on fasteners rise linearly with the absolute value of unbalanced superelevation. The peak frequency of fastener loads under unbalanced superelevation primarily falls within the ranges of 1.6–2.5 and 3.15–10 Hz. In contrast, straight sections or sections with balanced superelevation exhibit nearly linear load spectra. Compared to straight sections, curved sections show increases of up to 210% in peak fastener loads. The peak values of fastener loads, wheel–rail contact forces, vehicle accelerations, derailment coefficients, and wheel load reduction rates are mainly influenced by train speed, curve radius, and superelevation. Rail and slab displacements are additionally affected by spring stiffness. Fitting formulas derived for these variables provide a theoretical basis for optimizing the relationship among curve radius, train speed, superelevation, and steel‐spring stiffness, with the aim of improving both driving safety and vibration mitigation in railway systems. The peak vertical vibration acceleration of floor slabs generally increases with floor height, while peak lateral vibration acceleration shows no significant variation. Under balanced superelevation conditions, lateral acceleration spectra exceed those under straight‐track conditions, with notable differences in spectral shapes across floor levels. These results offer a theoretical foundation for optimizing track geometry and structural parameters in floating slab track systems, as well as a framework for predicting and assessing driving safety and structural vibration performance.

Deterioration mechanism of track geometry under lateral earthquakes and target ground motion-generated track alignment irregularity