<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Entire Interior Culvert Lining with Engineered Cementitious Composites</title>
      <link>https://trid.trb.org/View/2712609</link>
      <description><![CDATA[Corrugated metal pipes make up a significant portion of Virginia’s culvert inventory and are prone to deterioration from corrosion and abrasion, particularly at the invert. In early stages, invert rehabilitation techniques such as invert paving are sufficient. However, at advanced stages of distress, full interior repair becomes necessary. In such cases, spray-applied pipe lining offers a cost-effective alternative. For spray-applied pipe lining, achieving adequate fresh-state workability of the cementitious lining material is critical to ensure pumpability and sprayability while allowing the material to build the required thickness without slumping. Previous research developed engineered cementitious composites (ECCs) for invert repair, but these mixtures required a biaxial geogrid for stability and were not optimized for full interior lining. This study builds on prior work by evaluating ECCs as a trenchless rehabilitation solution for the entire interior culvert lining. Sprayable ECC mixtures were developed and tested in the laboratory using readily available materials, moderate fiber contents, and simplified compositions—including coal ash-based and slag cement-based systems—to enhance field practicality. In addition, a parametric study assessed existing liner thickness design methodologies, and field inspections were conducted to evaluate the performance of conventional cementitious liners and ECC-repaired culvert inverts. Laboratory results showed that ECCs incorporating air entrainment to leverage the Temporary High Initial Air Content concept achieved consistency suitable for pumpability, sprayability, and thickness buildup, while maintaining strength, deflection-hardening, and durability. The parametric study highlighted the need for more consistent design guidance and indicated that a 1- to 2-inch ECC liner thickness is sufficient to restore the structural capacity of most corrugated metal pipes. Field inspections revealed excellent performance of ECC-repaired inverts, with no cracking or debonding after up to 6.5 years in service, whereas conventional cementitious liners exhibited significant cracking within 3 to 4 years. Overall, these findings demonstrate that properly designed ECCs can provide superior crack control and durability, offering the Virginia Department of Transportation a robust and practical solution for trenchless corrugated metal pipe rehabilitation.]]></description>
      <pubDate>Sat, 13 Jun 2026 15:31:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2712609</guid>
    </item>
    <item>
      <title>Aggressive Rehabilitation of Gravel Roads with and without Enzymatic Soil Stabilizers</title>
      <link>https://trid.trb.org/View/2678366</link>
      <description><![CDATA[Granular-surfaced roads built on weak subgrades or subjected to high traffic loads are susceptible to damage as the original roadway cross-section settles and spreads over time, which also reduces the effectiveness of the drainage ditches for removing water from the surface. Such roads may require significant maintenance yet still fail to reach their intended performance and lifespan. Their performance can be restored through aggressive rehabilitation, involving retrieving subgrade material from the ditches to rebuild the cross-sectional roadway and ditch profiles, while compacting and optionally stabilizing the materials to improve strength and stability. In the present study, four test sections were aggressively rehabilitated on a granular-surfaced road at a military base in Johnston, Iowa. The road experienced decades of high traffic from heavy vehicles, causing it to widen significantly beyond its original width of approximately 26 ft. This flattened the ditches and inverted the crown which retained water instead of draining efficiently. For the first two test sections, after aggressively rebuilding the profile, the surfacing aggregates and several inches of subgrade were treated with a liquid soil stabilizer derived from plant-based organic enzymes to improve strength and stiffness. Smooth-tired compaction and sheepsfoot compaction were used for the first and second sections, respectively. The third and fourth sections were also aggressively rebuilt and constructed with smooth tire compaction in one section and sheepsfoot compaction in the other, but without enzymatic stabilizer. A fifth section without any rehabilitation, stabilization, or maintenance was also monitored and tested as a control. Field Automated Dynamic Cone Penetrometer (ADCP) and Light Weight Deflectometer (LWD) tests were performed to measure the strength and stiffness of the road sections at five different construction stages over two winter-spring cycles: (1) before construction, (2) after construction in fall 2023, (3) after thawing in spring 2024, (4) in fall 2024, and (5) after thawing in spring 2025. The abrasion resistance of the roadway surfacing aggregates was also evaluated under simulated traffic loading using a laboratory gyratory compaction device. The gradation of the coarse fraction of surface material was analyzed before and after gyratory compaction tests using the Iowa DOT’s rapid Camsizer. The results of the tests showed a general improvement in the mechanical properties of the enzyme-stabilized sections compared to the untreated sections. The field California Bearing Ratio (CBR) values determined from correlations to ADCP tests and field LWD stiffness values both increased significantly over all time stages due to enzymatic stabilization, which can be attributed to increased cohesive bonding between the soil particles. These findings highlight the ability of aggressive rehabilitation to restore drainage ability, as well as the practical applicability and potential use of enzymatic chemical stabilizers to improve the performance of granular-surfaced roads.]]></description>
      <pubDate>Fri, 12 Jun 2026 15:59:27 GMT</pubDate>
      <guid>https://trid.trb.org/View/2678366</guid>
    </item>
    <item>
      <title>UHPC Roadmap: Advancing the Use of Ultra-High Performance Concrete (UHPC) in Bridge Applications ​ ​</title>
      <link>https://trid.trb.org/View/2712044</link>
      <description><![CDATA[In the early 2000s, a new class of cementitious materials—known as Ultra-High Performance Concrete (UHPC)—emerged. UHPC is a highly durable material characterized by low or discontinuous porosity, which significantly reduces water and chemical ingress. These properties help mitigate concrete deterioration, thereby supporting the FHWA’s goal to “Get in, Get out, and Stay out” in bridge construction and lowering maintenance costs.

With support from the Iowa Highway Research Board (IHRB) and the Iowa DOT, several projects have explored the application of UHPC in various bridge components over the past two decades. These applications include girders, waffle decks, piles, and deck overlays. Additionally, UHPC has been used to connect prefabricated elements and to repair damaged structural members.

Recognizing the significant potential of UHPC in both the rehabilitation of existing bridges and the design of new bridges in the United States—and building on its involvement in earlier projects—a group of interested individuals came together to establish an organization that would bring together academia, government agencies, and industry to promote the use of UHPC in bridge applications. Following the 2019 UHPC International Symposium, an initiative to advance the understanding and application of UHPC was launched at Iowa State University. This initiative, known as the UHPC Initiative (or UHPCi), is dedicated to advancing UHPC for civil infrastructure by supporting both fundamental and applied research. It also maintains an active program of knowledge and technology transfer among academic institutions and partner organizations. The mission of UHPCi is to provide a collaborative forum that fosters the sharing of knowledge and promotes research, innovation, and the development of specifications to increase the understanding and use of UHPC.

Despite the progress achieved, further work is needed to advance the use of Ultra-High Performance Concrete (UHPC) in bridge applications. Key areas for advancement include identifying the most suitable applications, reducing material and construction costs, and enhancing bridge longevity with minimal maintenance. Recent inspections conducted by both the Iowa DOT and personnel on past UHPC projects have revealed some concerns, which should also be investigated as part of this ongoing effort. A persistent challenge is the high cost of UHPC. Promising strategies to address this include reducing the metallic fiber content or partially substituting metallic fibers with non-metallic alternatives. These approaches could be explored on an application-specific basis to optimize performance and cost-effectiveness.

Given the aforementioned challenges and opportunities, this pooled fund project—conducted in collaboration with the UHPCi—aims to consolidate current knowledge and develop a strategic roadmap to enhance the use of UHPC in both bridge rehabilitation and new bridge construction. The effort will focus on identifying the most promising applications (e.g., bridge decks), where UHPC can offer the greatest impact.]]></description>
      <pubDate>Mon, 08 Jun 2026 11:32:59 GMT</pubDate>
      <guid>https://trid.trb.org/View/2712044</guid>
    </item>
    <item>
      <title>Impact of Tamping on Repeated Ballast Settlement</title>
      <link>https://trid.trb.org/View/2675872</link>
      <description><![CDATA[It has been confirmed that track irregularity gradually returns to its original shape after ballast tamping on ballasted track even under the same track and support structure conditions. However, the details of this mechanism are not yet clear. Therefore, we surveyed an actual situation using track inspection data for this phenomenon. In addition, we performed tests with small-model, discontinuum analysis for ballast density after ballast tamping and cyclic loading tests, to reveal the mechanism of reversion in settlement after and before ballast tamping.]]></description>
      <pubDate>Fri, 05 Jun 2026 16:41:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2675872</guid>
    </item>
    <item>
      <title>Monitoring Rail Bed Infrastructure Using Wireless Passive Sensing</title>
      <link>https://trid.trb.org/View/2663117</link>
      <description><![CDATA[Railroad ballast plays a critical role in distributing train loads, maintaining track stability, and ensuring proper drainage. Over time, intrusion of fines and moisture leads to ballast fouling, which can compromise track structural integrity and service life. Early detection of ballast fouling is therefore essential for effective maintenance and extending the operational life of rail infrastructure. This study investigated the use of passive harmonic transponders as a low-cost, sustainable, and long-lasting sensing solution for monitoring ballast conditions. Laboratory experiments were conducted using clean and fouled ballast under varying moisture levels and transponder embedment depths. The harmonic transponder is unique in that it returns a signal at twice the frequency of that which interrogates it, allowing clear separation of the transponder response from background reflections and noise. Two types of harmonic transponders, one is commercially available sensing device (RECCO, operated at fundamental frequency 890 MHz) and another university developed sensing device (developed by Czech Technical University, operated at 1.17 GHz), were tested to evaluate signal responses under different conditions. The results indicate that moisture content and embedment depth are the primary factors influencing signal attenuation, while fouling level plays a secondary role. The transponders operated at lower frequency demonstrated more stable performance, whereas the device operated at higher frequency was more sensitive to ballast heterogeneity and moisture fluctuations. The findings suggest that embedding passive harmonic transponders during new rail-bed construction can provide a built-in monitoring system capable of early fouling detection. Interrogation of the embedded transponders can be performed using portable units, moving trains, or drones, offering a flexible and non-destructive approach. Their low-cost and passive design eliminates the need for batteries or continuous power supply, reducing both maintenance requirements and environmental impact. This energy-efficient operation enables long-term field deployment with minimal resource consumption. As a result, harmonic transponders represent a practical and sustainable solution for continuous monitoring of fouling in railway ballast.]]></description>
      <pubDate>Thu, 04 Jun 2026 10:58:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2663117</guid>
    </item>
    <item>
      <title>Multi-objective resilience-based optimal prioritisation of ageing bridge networks under concurrent multiple hazards in a changing climate</title>
      <link>https://trid.trb.org/View/2708286</link>
      <description><![CDATA[Bridges are critical components of transportation networks and are highly vulnerable to various hazards throughout their lifespan. Natural disasters such as earthquake sequences and riverine floods create serious risks, while climate change is making these threats even more severe. In fact, climate change is increasing the frequency and intensity of extreme weather events, leading to more severe flooding and accelerated deterioration of structures, with a detrimental impact on the life-cycle performance and resilience of bridges and bridge networks. It is therefore essential to ensure adequate pre- and post-event functionality, particularly for traffic flow and road connectivity. To this purpose, rational prioritisation of rehabilitation activities is key for optimising limited resources and reducing both direct and indirect economic losses. This paper presents a probabilistic, resilience-based framework aimed at prioritising retrofit and restoration interventions on ageing bridges and bridge networks facing multiple hazards, including seismic mainshock and aftershocks, flooding, and corrosion under climate change. The proposed approach employs multi-objective optimisation to identify the most effective pre- and post- event activities that enhance resilience while minimising both direct and indirect costs. The application to a bridge transportation network illustrates how the proposed framework allows to addressing strategic rehabilitation planning and improving resilience while ensuring cost-efficiency.]]></description>
      <pubDate>Wed, 03 Jun 2026 09:07:20 GMT</pubDate>
      <guid>https://trid.trb.org/View/2708286</guid>
    </item>
    <item>
      <title>An Alternative Approach to Track Settlement Prediction</title>
      <link>https://trid.trb.org/View/2113174</link>
      <description><![CDATA[Many empirical equations have been formulated in an attempt to model the settlement of ballasted railway track at an individual sleeper. Some equations have been used with vehicle track interaction models (VTI) to predict the development of differential settlement along the track iteratively and hence maintenance requirements, over potentially millions of cycles of load. For settlement equations to be suitable in such simulations, they require as an input a VTI model output that varies along the track, such as the force on the sleeper or the current resilient deflection range. For computational economy, these VTI simulations are usually run in large steps with the settlement predicted forward over many cycles. There is, however, no one generally applicable settlement equation, and it remains unclear whether the loss of accuracy that ensues from stepping the VTI analyses is acceptable. A realistic settlement equation needs to incorporate both stress- and load history-dependent behaviour. This paper proposes a new settlement model that allows for stress history and has the potential to be applied at every cycle within an iterative VTI simulation. The ballast layer is modelled by combining a nonlinear visco-elastic element to simulate the resilient response with a plastic-hardening element for permanent settlement. This leads to the calculation of permanent settlement without recourse to an explicit empirical equation. The parameters used in the model are determined using data from cyclic laboratory tests on a single sleeper. The effect of different loading histories on the model is considered.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113174</guid>
    </item>
    <item>
      <title>The Use of Microporous Membranes to Address Mud Pumping—UK Experience</title>
      <link>https://trid.trb.org/View/2113169</link>
      <description><![CDATA[The UK has one of the most complex, most heavily trafficked rail networks in the world. In most areas, there is little time available for maintenance or track renewal; as a result, the installation of a completely new roadbed is seldom justifiable. Over the last 30 years, the need for time-consuming treatments requiring large volumes of excavation and imported sub-ballast has been gradually eliminated by maximising the use of geosynthetics to enhance the mechanical performance of the existing sub-ballast and subgrade. However, prior to the introduction of microporous membranes there was no treatment for a severe subgrade erosion problem that did not require installation of a new granular layer before placing new ballast. The paper summarises early roadbed treatments, subsequent research work on geosynthetics and the development of a suite of geosynthetic treatments. These would mitigate problems with existing sub-ballast or subgrade before placing new ballast. The development of microporous membranes is then described. Finally, the paper summarises the current methodology used by network rail (NR) to select an appropriate treatment for any given site.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113169</guid>
    </item>
    <item>
      <title>A Deep Investigation into the Mechanisms and Factors Producing Mud Pumping</title>
      <link>https://trid.trb.org/View/2113156</link>
      <description><![CDATA[Mud pumping is a common issue facing all railroads in which wet fines within the ballast pump up around the ties and onto the surface of the track during train loading. This often corresponds to poor drainage, loss in track geometry, reduced ballast strength and stiffness, and in the worst case leading to ballast failure. Despite the prevalence of this problem, the mechanisms behind mud pumping and the factors influencing it are not fully understood, although past investigations have determined that fines and water in ballast, as well as repeated dynamic wheel loads, need to act together to produce mud pumping. An improved understanding is to how it will allow for better prediction of mud pumping and the negative effects associated with it, along with improved maintenance planning and techniques. Transportation Technology Center, Inc. (TTCI) has further reviewed and investigated mud pumping situations from both its “rainy section” test zone at the Facility for Accelerated Service Testing (FAST) near Pueblo, Colorado, and multiple revenue service locations in attempt to understand the underlying mechanisms behind mud pumping. The rainy section at FAST and some revenue service sites suggested that some (and possibly most) mud pumping situations are purely surficial, originating from the wet regions just around the ties. However, other mud pumping situations have shown a seasonal or perched water table below the bottom of the ties that appears to cause moisture and fines to pump up to the surface from the subgrade or lower ballast layer. TTCI plans to continue to investigate different mud pumping situations with the end goals of developing ballast maintenance guidelines and improving track substructure-induced track geometry degradation forecasting models.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113156</guid>
    </item>
    <item>
      <title>Dynamic Behavior Modeling of Full-Scale High-Speed Ballasted Track Using Discrete Element Method</title>
      <link>https://trid.trb.org/View/2113137</link>
      <description><![CDATA[Ballast layer dynamic behavior is critically important for railway track design and maintenance optimization. This paper presents findings on crosstie and ballast particle dynamic responses obtained from: (i) laboratory tests conducted at Zhejiang University innovative high-speed rail tester (ZJU-iHSRT) and (ii) discrete element method (DEM) simulations using algorithms with newly featured parallel computing capability developed at UIUC. Overall, more than 170,000 ballast particles and eight crossties were assembled in the DEM model. A proportional integral derivative (PID) controller was utilized to ensure realistic dynamic loads applied on crossties at three train speeds: (1) 108 km/h; (2) 252 km/h; (3) 300 km/h. Crosstie vibration velocities predicted using the DEM model matched closely with measurements from laboratory tests both in trends and in magnitudes. With Fourier transformation and Butterworth filter techniques implemented on ballast particle vibration velocities captured in the DEM model, inherent signal noise could be reduced, and as a result, the predicted ballast particle vibration trends matched closely with laboratory sensor measurements. However, individual ballast particle vibration magnitudes predicted by the DEM simulations revealed certain discrepancies with the measurements since velocity sensors used in the experiment only recorded vibration responses of an assembly of ballast particles. Further studies are necessary to reveal more detailed findings.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113137</guid>
    </item>
    <item>
      <title>Effect of Degraded Subgrade Stiffness on Rail Geometry and Train Vibration in High-Speed Railways</title>
      <link>https://trid.trb.org/View/2113129</link>
      <description><![CDATA[The degradation of railway subgrade stiffness will cause the acceleration of track degradation and deteriorate ride comfort and safety. It is hard to foresee these problems because both track substructure and subgrade are not visible to inspect. To study the influence of subgrade stiffness degradation on high-speed train and track system, 3D FEM analyses of the Fuxing trains and ballastless tracks were conducted to simulate different degradation conditions of roadbed stiffness. Firstly, the numerical model was verified by the field measurement of vibration velocity at the concrete base in the Beijing–Tianjin high-speed railway. Then, the influences of the degradation of roadbed stiffness on the rail displacements and train dynamics were analyzed. It is found that since the ballastless track had quite high rigidity, the uneven degradation of roadbed stiffness did not result in obvious additional increase of rail displacement. However, the uneven degradation of roadbed stiffness resulted in larger vibrations of train wheels. Particularly, the wheel accelerations were more sensitive to the uneven degradation of roadbed modulus amplitudes than the wavelengths.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113129</guid>
    </item>
    <item>
      <title>Mud Pumping in Ballastless Slab Track of High-Speed Railway and Its Remediation</title>
      <link>https://trid.trb.org/View/2113128</link>
      <description><![CDATA[Mud pumping is an increasing subgrade distress in ballastless slab track of high-speed railway, which heavily affects the driving comfort and even threatens the driving safety. However, the existing researches mostly focused on the water-induced distress in ballasted railway, and little attention was paid on the mud pumping in ballastless slab track. In this paper, both in-situ investigation and discussion of the conventional grouting remediation method were carried out. It can be observed that mud pumping mainly occurred at the expansion joint located at the ends of concrete base and then expanded to both sides of the expansion joint from 1–2 m. The defects in railway structure and the standing water stored in the roadbed layer were the two main factors contributing to mud pumping. In addition, based on the traditional chemical glue injection (CGI) remediation method (the shallow grouting repair method), a modified polyurethane grouting remediation method (the deep grouting repair method) was put forward to treat mud pumping in ballastless slab track.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113128</guid>
    </item>
    <item>
      <title>Ballasted Track Maintenance Modelling Using DEM</title>
      <link>https://trid.trb.org/View/2113123</link>
      <description><![CDATA[The ballast layer is a crucial component of railway tracks and it is hence essential to maintain it using adequate processes like tamping and stabilization. These will ensure that the density of the ballast layer is high enough to avoid shearing and settlement of the track under traffic. Ballasted tracks settle unevenly under the passage of trains. These geometrical defects are corrected by tamping which consists of lifting individually the sleepers and compacting the ballast underneath using vibrating tines. After tamping, the ballast layer is not homogeneous in terms of density along the track and requires stabilization before being commercially operational. This stabilization is performed either by regular trains at lower speeds for a given period hindering commercial operations, dynamic stabilization, or crib compaction. All these processes rely on vibrating the ballast layer using different approaches and have mainly been based on empirical observations. This paper describes an analysis of these ballasted track maintenance processes and their optimization using the discrete element numerical approach. This approach considers a granular material as an assembly of objects interacting through a specific contact law. In the present study, the code called LMGC90 has been used. The study includes a comparison of the processes in terms of ability to compact the ballast layer and lateral mechanical resistance of the track and their optimization. The final purpose of the project is to be able to specify optimal functioning parameters for all these processes.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2113123</guid>
    </item>
    <item>
      <title>YOLOv8DTL: A Deep Transfer Learning Model for Few-Shot Rail Abrasion Detection</title>
      <link>https://trid.trb.org/View/2658869</link>
      <description><![CDATA[Rapid movement of the wheels on defective tracks and high-frequency friction collisions cause vibration between train and rail track, which greatly damages the lifespan of train components and is a significant cause of train derailments. Timely and accurate detection of rail abrasions is of great significance for ensuring the safety of railway operations. Deep learning based-automatic rail abrasion detection methods face the challenge of having fewer samples. Thus, a deep transfer learning framework based on improved YOLOv8 models is developed to detect rail abrasions to address the issue of few-shot learning. Then, deformable convolution networks (DCNs), convolutional block attention module (CBAM) and a new intersection over union (IoU) loss are introduced to improve the detection performance. The ablation experiments show that it effectively reduces the rate of missed and false abrasion detections. By comparing with the existing detection methods, the developed YOLOv8DTL method has higher precision, recall, and average precision under different abrasion size thresholds, indicating that it is more adaptable to detection tasks with different abrasion sizes. It also has the best robustness, maintaining a high level of detection efficiency.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:35 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658869</guid>
    </item>
    <item>
      <title>Accurate and Fast Quantitative Detection of Rail Corrugation Driven by Deep Learning Algorithms Using Adaptive Targeted Loss and Directed Metric Functions</title>
      <link>https://trid.trb.org/View/2658841</link>
      <description><![CDATA[Rail corrugation is a common rail surface defect in wheel–rail system and can significantly reduce the running quality of vehicles. Accurate and fast detection of corrugation is of great importance for uncovering its mechanism, predicting the evolution and planning maintenance. Traditional physical model-driven detection approaches rely on high-fidelity simulations and are poorly automated. The popular data-driven methods have good availability. However, they ignore the power of expert experience, resulting in limited accuracy. This paper presents a quantitative model for detecting rail corrugation based on measurement data and expert knowledge using deep learning (DL) algorithms. First, a data quality improvement framework is proposed to overcome the problem that conventional DL models may produce incorrect results when trained on low-quality datasets. Then, the characteristics of rail corrugation are deeply investigated and adaptive targeted loss and directed metric functions are developed according to the individual samples to build a detection model with better global feature regression and local detail convergence. Finally, the effectiveness of the proposed method is verified using on-site measurement data from a commercial metro line. Comparative analyses and ablation studies demonstrate the superiority of targeted loss and directed metric in terms of more accurate detection and scientific evaluation of rail corrugation roughness, respectively.]]></description>
      <pubDate>Thu, 28 May 2026 17:09:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2658841</guid>
    </item>
  </channel>
</rss>