<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>Harnessing Machine Learning for Railway Damage Assessment and Prediction</title>
      <link>https://trid.trb.org/View/2678439</link>
      <description><![CDATA[Railway systems in coastal regions face distinct challenges due to environmental factors such as coastal erosion, hurricanes, and rising sea levels. Coastal erosion undermines railway embankments, resulting in structural instability and heightened maintenance costs. Hurricanes and storm surge present further risks, with high winds and flooding threatening to damage rail infrastructure, disrupt services, and compromise public safety. To address these vulnerabilities, railway systems in these areas require advanced engineering solutions capable of mitigating the effects of extreme weather events while ensuring operational continuity. This thesis seeks to leverage machine learning techniques to assess and quantify damage to railway infrastructure following major storm events, providing valuable insights to inform the design and maintenance of coastal railway systems. By collecting and processing data on railway characteristics, storm events, and resulting damage, a predictive machine learning model was developed to estimate future storm-related damage. The model achieved an accuracy of approximately 84%, demonstrating its potential for practical application. Additionally, three hypothetical scenarios were developed to evaluate the model’s performance by varying key input parameters. The findings of this study offer critical guidance for railway stakeholders in making informed decisions regarding the reinforcement of coastal railways and pave the way for further research and model refinement.]]></description>
      <pubDate>Fri, 02 Oct 2026 16:48:00 GMT</pubDate>
      <guid>https://trid.trb.org/View/2678439</guid>
    </item>
    <item>
      <title>Assessment of Expanded Clay Aggregate for Use in Railroad Embankments</title>
      <link>https://trid.trb.org/View/2678490</link>
      <description><![CDATA[Ground settlement is a major challenge in railroad embankment construction and widening, threatening both track stability and nearby infrastructure. Settlement arises from two mechanisms: immediate elastic compression of soils and longer-term consolidation. To mitigate these risks, lightweight aggregates (LWAs), such as expanded clay, offer a promising solution by reducing embankment load and associated settlement. Expanded clay is nearly half the weight of conventional aggregates, with high internal friction angle and excellent drainage, enhancing stability and reducing hydrostatic pressure. However, its long-term performance under cyclic train loading remains uncertain. To address this, a laboratory testing program, including large-scale Cyclic Direct Simple Shear (CDSS) tests, was developed to simulate railroad loading conditions and evaluate the durability and deformation behavior of expanded clay. This study aims to validate LWAs as a viable alternative for minimizing settlement without compromising long-term track performance.]]></description>
      <pubDate>Fri, 02 Oct 2026 15:56:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2678490</guid>
    </item>
    <item>
      <title>Stabilizing a Landslide to Preserve the Historic Irishtown Bend in Cleveland, OH</title>
      <link>https://trid.trb.org/View/2678375</link>
      <description><![CDATA[The Irishtown Bend Stabilization Project is a major infrastructure project in Cleveland, Ohio, that involves stabilization of a massive landslide to restore a historic neighborhood to public use and prevent a slope failure that threatens to impair navigation along the economically vital Cuyahoga River shipping channel at the toe of slope. Urban development and fill placement in the 1960s added load to the top of the Irishtown Bend hillside. The added load reactivated ancient slide planes at a depth of more than 30 m (100 ft) and triggered large, on-going slope movements including a major head scarp at the top of slope. The slide is retrogressive today with local slope failures at the hillside toe occurring due to on-going erosion from river action and turbulence caused by ship traffic. This paper discusses the geologic history of the Irishtown Bend, site development, hillside instability, and the on-going slope movement. The results of a 2019 subsurface investigation and slope movement monitoring program are described with focus on defining the landslide failure geometry and developing peak and residual friction angles in the glacial lacustrine clay for use in stability assessment. This paper also discusses the results of back analysis of slope stability and assessment of the necessary stabilization measures to achieve target safety factors. Finally, design and construction of long-term stabilization measures are presented including major earthwork to facilitate hillside unloading and installation of a permanent anchored bulkhead consisting of a steel sheet pile and pipe pile combination wall braced by tieback anchors extending to shale bedrock.]]></description>
      <pubDate>Fri, 02 Oct 2026 15:56:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2678375</guid>
    </item>
    <item>
      <title>Coupled Multichannel Analysis of Surface Waves and Electrical Resistivity Imaging Methods for Assessing Sinkhole-Impacted Railroad Embankment: Case Study</title>
      <link>https://trid.trb.org/View/2782943</link>
      <description><![CDATA[This study characterizes a sinkhole-impacted railroad embankment site through a comprehensive geophysical investigation encompassing eight electrical resistivity imaging (ERI) lines, four multichannel analysis of surface waves (MASW) lines, and three LiDAR datasets. Geophysical data from the coupled ERI and MASW at the south embankment mid-slope and toe revealed low resistivity (<50 Ω·m) and low shear-wave velocity (Vs) to depths of 25 to 30 ft, indicating moisture-weakened or void-prone zones. Increased Vs values beyond a 30 ft depth indicated stiffer soil or bedrock. Confined moisture pockets were highlighted along the sinkhole parallel to the slope’s toe. Widespread low-resistivity zones and low Vs were found at the mid-slope and toe to 30 ft, suggesting saturated, softened zones in those areas. Low resistivity and low Vs were found at several locations in the crest, railway track, and subgrade zones between 10 and 40 ft and at ∼240 ft lateral distances along the investigation lines at the crest of the railroad embankment, mostly within 15- to 30-ft depths. Low-resistivity and low-Vs zones beneath and adjacent to known sinkholes at the north slope toe indicated saturated voids and potential zones of instability. Resistivity and Vs variations indicated active seepage, hydraulically active zones, and potential sinkhole depths on both sides of the embankment. The coupled MASW and ERI investigation method revealed subsurface anomalies associated with sinkhole formation, narrowing down the critical areas for potential interventions. This noninvasive approach enabled rapid and cost-effective screening of railroad assets for prioritized proactive maintenance, particularly following extreme weather events.]]></description>
      <pubDate>Tue, 29 Sep 2026 09:52:28 GMT</pubDate>
      <guid>https://trid.trb.org/View/2782943</guid>
    </item>
    <item>
      <title>Transient Deformation Analysis of a Peat Foundation Under Heavy-Haul Railway Loading</title>
      <link>https://trid.trb.org/View/2752207</link>
      <description><![CDATA[A three-dimensional finite-difference model was developed to identify the mechanisms governing lateral deformation in a peat-supported heavy-haul railway embankment at the Anzac site, Alberta, Canada. ShapeAccelArray measurements showed that the maximum lateral deformation occurred near the peat–silty clay interface and could not be reproduced using a conventional layered representation of the foundation. Site-specific peat properties derived from laboratory testing were used in the moving-load analyses, while sensitivity analyses evaluated the effects of boundary conditions, stiffness contrast, interface behavior, and loading configuration. The conventional model placed the maximum deformation within the peat, whereas a thin layer of weak material at the peat–silty clay interface reproduced the measured depth-dependent lateral deformation profiles. Static point-load analyses captured the dominant transverse response but underestimated longitudinal deformation. Moving-load simulations reproduced the measured deformation trends parallel and perpendicular to the track. Maximum shear strain increments were concentrated near the peat–silty clay interface and attenuated within the underlying silty clay. The results showed that localized interface behavior, foundation stiffness contrast, and load progression governed the observed lateral deformation response.]]></description>
      <pubDate>Mon, 28 Sep 2026 08:37:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2752207</guid>
    </item>
    <item>
      <title>Automated Finite Element Modeling and Predictive Assessment of Deformation in HSR Bridge–Embankment Transition Zones on Soft Soil</title>
      <link>https://trid.trb.org/View/2752205</link>
      <description><![CDATA[The construction of high-speed railway (HSR) bridges and approach embankments over thick soft soil deposits presents significant geotechnical challenges, including excessive settlement, differential deformation, and lateral displacement. This study introduces an integrated numerical–predictive framework for evaluating the performance of HSR bridge abutments and transition embankments on soft ground, with particular emphasis on the Mekong Delta region of Vietnam. Two-dimensional finite element analyses were conducted to examine the influence of prestressed high-strength concrete pile length and diameter, braking force, and vertical train load on the settlement and lateral displacement of the abutment. The preliminary results demonstrated that the combined pile–geocell reinforcement system significantly reduced both total and differential settlements while enhancing deformation capacity within the transition zone. Numerical simulations were subsequently used to develop predictive equations for settlement and lateral displacement using Gene Expression Programming. The developed models demonstrated satisfactory predictive capability and successfully captured the nonlinear interactions among loading and reinforcement variables. The parametric study further indicated that emergency braking increased the lateral displacement, whereas longer piles substantially improved the deformation control. Overall, the proposed framework provided practical guidance for deformation assessment and preliminary design of HSR bridge–embankment transition systems over thick soft soil deposits.]]></description>
      <pubDate>Mon, 28 Sep 2026 08:37:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2752205</guid>
    </item>
    <item>
      <title>Contextualising the impact of street slopes on pedestrian accessibility for sustainable mobility</title>
      <link>https://trid.trb.org/View/2778178</link>
      <description><![CDATA[Proximity-based planning has become a central paradigm in transport policy, guiding the design of sustainable, accessible, and livable cities. This shift has driven the development of various methods for assessing pedestrian accessibility, reflecting the central role of walking as the most sustainable mode. Yet, the impact of street slopes on accessibility is often neglected and remains poorly understood. This study identifies two systematic biases: a slope-speed bias, caused by assuming flat terrain, and a slope-preference bias, reflecting behavioural aversion to slopes. We compare three models: (a) slope-agnostic, (b) slope-adjusted with gradient-dependent speeds, and (c) slope-preference incorporating walking preferences across walking propensity groups. Applied in Athens, Greece, results show that slope-agnostic models overestimate accessibility by approximately 6–8% on average. Accounting for walking preferences introduces an additional 33–40% reduction for low-propensity walkers across destination types. The proposed framework offers a transferable approach for integrating terrain sensitivity into accessibility analysis.]]></description>
      <pubDate>Mon, 28 Sep 2026 08:37:08 GMT</pubDate>
      <guid>https://trid.trb.org/View/2778178</guid>
    </item>
    <item>
      <title>Risk Assessment-Based Premium Rate Determination for Highway Slope Insurance: Method and Application</title>
      <link>https://trid.trb.org/View/2780267</link>
      <description><![CDATA[Premium rate setting is a key technical issue for ensuring the sustainability of highway property insurance. A risk assessment-based premium rate-setting method for natural disaster insurance of highway slopes is proposed, which achieves a long-time balance between insurance claims and premiums and promotes the sustainability of highway insurance. The core of this method lies in quantifying the conditional property losses of highway slopes under natural disasters through risk assessment. By combining this with historical insurance claims data, the probability of disaster losses in the region is estimated, thereby enabling the derivation of appropriate insurance premiums. This method was verified using highway insurance cases in Ningbo. Six slopes with different technical and environmental conditions were selected to validate the calculated slope conditional loss. The disaster loss probability of the slopes in Ningbo is derived from four years of risk assessments and insurance compensation data for the county road insurance in Haishu District, Ningbo. Introducing this disaster loss probability, performing risk assessments and conditional loss estimations for all hazard-prone slopes, a reasonable premium rate for 1,384 km of national and provincial highways in Ningbo is calculated, and the differentiated rates for each sub-region are also obtained.]]></description>
      <pubDate>Wed, 23 Sep 2026 15:57:02 GMT</pubDate>
      <guid>https://trid.trb.org/View/2780267</guid>
    </item>
    <item>
      <title>Moisture evolution and instability mechanisms in high-fill loess embankment slopes subject to water infiltration from accumulated runoff following berm erosion</title>
      <link>https://trid.trb.org/View/2709975</link>
      <description><![CDATA[Localized water accumulation and siltation significantly compromise the structural integrity of loess embankments and slopes, leading to slope failure and differential settlement—posing severe safety risks to highway operations. This study investigates the mechanisms of slope instability and uneven settlement caused by localized rainwater infiltration at the slope berm of a high-fill loess embankment under intermittent extreme rainfall conditions. The results show that the infiltration process is influenced by both hydraulic pressure and ambient temperature. Accumulated water at the slope berm can initiate slope collapse and cracking, with prolonged rainfall and localized waterlogging being the primary contributors to hydraulic degradation of the embankment. The failure evolution can be divided into five distinct stages: initial infiltration, increased slope mass, local subsidence, tensile cracking, and eventual embankment collapse. Water tends to migrate downslope toward the toe, serving as a critical trigger for slope instability and landslides. During infiltration phase II (with cumulative water injection exceeding 36 L), the saturated mass moisture content reaches 32%. For the embankment body, settlement at each monitoring point exhibits a quadratic relationship with increasing mass moisture content, whereas settlement at the embankment-foundation interface follows a linear trend. The pore water pressure fluctuates over time in a wavelike pattern, ranging from a maximum of 2.21 kPa to a minimum of −0.23 kPa. Monitoring results reveal that most settlement occurs after soil saturation, with a ratio of S(before saturation):S(after saturation) ranging from approximately 1.06 to 2.25. Driven by the saturation of the soil, the “sliding collapse” failure mode of the embankment is eventually induced. The instability mechanisms elucidated in this study provide an academic reference for the construction, hydraulic design, and early-warning frameworks of high-fill loess embankments.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709975</guid>
    </item>
    <item>
      <title>Strength-deformation and critical state of coarse-grained soil considering particle shape and particle breakage</title>
      <link>https://trid.trb.org/View/2709916</link>
      <description><![CDATA[Coarse-grained soils are extensively employed as primary construction materials in civil infrastructure, including high earth-rock dams and railway embankments, where long-term serviceability is of paramount importance. Particle shape and particle breakage are two key factors that influence the mechanical behavior of coarse-grained soils. In this study, a series of consolidated drained triaxial tests were performed to evaluate the effects of shape and breakage on the mechanical and critical state behavior of coarse-grained soils. The results show that, for a given particle shape parameter of overall regularity, deviatoric stress, peak strength, and relative breakage index increase with confining pressure, while the peak friction angle decreases. An increase in overall regularity under a fixed confining pressure increases the deviatoric stress, volumetric strain, peak strength, and peak friction angle, and is accompanied by a reduction in the relative breakage index. The critical state lines rotate counterclockwise in the p′-q plane (i.e., stress plane). And in the e-p′/paξ plane (i.e., deformation plane), they shift downward and rotate counterclockwise as overall regularity with increase (for ξ = 0.7). Based on these observations, a three-dimensional critical state surface was developed to incorporate the effects of particle shape and particle breakage, and the proposed surface provides a good representation of the experimental behavior.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709916</guid>
    </item>
    <item>
      <title>Prediction of freeze–thaw interface depth in a permafrost highway embankment using variational mode decomposition and a CNN–LSTM hybrid model</title>
      <link>https://trid.trb.org/View/2710205</link>
      <description><![CDATA[Accurate prediction of freeze–thaw interface depth (FTID) is critical for assessing the thermal stability of highway embankments in permafrost regions. Monitoring-derived time series are nonlinear, nonstationary, and dominated by mixed-scale variability. This study developed a VMD–CNN–LSTM framework, where variational mode decomposition (VMD) partitions each meteorological predictor into trend, seasonal, and residual components; a one-dimensional convolutional neural network (CNN) extracts the localized nonlinear patterns; and a long short-term memory (LSTM) network learns temporal persistence and lagged responses. The framework was evaluated using semi-monthly meteorological forcing and borehole ground temperature data (2004–2021) from the Xidatan monitoring section (K2887 + 500) along the Qinghai–Xizang Highway, with a strictly chronological split (first 90% for training/validation; last 10% for testing). At the road centerline, the proposed model outperformed LightGBM, XGBoost, and LSTM-family baselines (R² = 0.965; MAE = 0.037 m). Generalization tests at the left shoulder and natural ground positions yielded R² ≥ 0.952. The ablation analysis indicated that VMD drove most of the error reduction by isolating low-frequency trend/seasonal signals from high-frequency disturbances. The workflow can support the monitoring-based diagnosis of embankment thermal states and early warnings of abnormal thaw development, thereby providing a critical precursor variable for subsequent thaw settlement assessment and maintenance planning.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2710205</guid>
    </item>
    <item>
      <title>Real-time soil moisture content detection during compaction based on near-infrared technology and thermo-hydraulic migration simulation</title>
      <link>https://trid.trb.org/View/2710191</link>
      <description><![CDATA[Ensuring the compaction quality of soil materials is essential for controlling post-construction settlement in dams and high-fill embankments. As moisture content is a key factor affecting soil compaction quality, its rapid and accurate detection during rolling is of great practical importance. However, conventional methods, such as oven drying and alcohol combustion, rely on limited-point sampling and cannot provide rapid full-field detection of in situ moisture content, which varies significantly with time, material source, and paving position. To address this limitation, this study proposed a real-time method for detecting the overall soil moisture content during rolling based on near-infrared (NIR) technology. A compactor-mounted real-time detection device was first developed to measure the soil surface moisture content by integrating intelligent compaction technology with NIR sensing. A thermo-hydraulic coupled migration model considering air temperature, wind speed, and light intensity was established to relate the NIR-detected surface moisture content to moisture content at different depths within the compacted layer. On this basis, an efficient surrogate model, namely a goose algorithm-optimized back-propagation neural network (GOA-BPNN), was constructed to rapidly predict soil moisture content at different depths. A calculation method was further proposed to estimate the overall moisture content within the paving layer. Finally, a real-time soil moisture detection system was developed to achieve rapid and comprehensive detection of the overall soil moisture content during rolling. Field application results show that the mean absolute error between the detected and actual overall soil moisture content is 0.86%, demonstrating high detection accuracy. The proposed method provides an efficient approach for real-time detection of the overall soil moisture content during rolling and offers more precise and comprehensive moisture information for intelligent compaction-based assessment of soil compaction quality.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2710191</guid>
    </item>
    <item>
      <title>Cooling performance of a self-regulating mechanical ventilation system: A case study for Qinghai-Tibet expressway Tuotuohe pilot section</title>
      <link>https://trid.trb.org/View/2709852</link>
      <description><![CDATA[The thermal stability of embankments in permafrost regions is essential for the long-term safety and serviceability of transportation infrastructure. Conventional mechanically ventilated embankments can regulate the thermal regime of permafrost subgrades. However, fixed-airflow operation has limited adaptability to seasonal and interannual climatic variability, resulting in insufficient control of cold-energy input. To overcome these limitations, this study proposed a half-monthly dynamic cold-energy self-regulation strategy that combines planned cold-energy allocation with cumulative-deviation feedback control. A three-dimensional numerical model of a Qinghai-Tibet Expressway pilot section was established to evaluate the performance of the proposed strategy under mean climate, warm-year/cold-year fluctuations, and extreme warm climate conditions. Results indicate that: 1) Under the mean climate, a basic airflow speed of 1.97 m/s achieved cold-energy balance and maintained a stable artificial permafrost table beneath the embankment; 2) the proposed strategy showed good adaptability under warm-year, cold-year, and extreme warm-year conditions, reducing cold-energy deficit and improving the thermal state in warm years, suppressing excessive cooling and lowering electricity consumption from 23.98 to 19.23 kWh in cold years, and providing better practical economy than the extreme-compensation scheme under extreme warm conditions; 3) under the continuous climatic sequence, the self-regulation scheme maintained a shallower permafrost table and lower cumulative cold-energy mismatch than the basic scheme and the unmodified subgrade, indicating better long-term thermal resilience under sustained climatic variability. The proposed strategy provides a practical basis for setting airflow design values and operation rules for mechanically ventilated embankments on high-altitude permafrost highways.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709852</guid>
    </item>
    <item>
      <title>Analysis of the cooling effect of duct-ventilated embankments with different duct spacings in high-temperature permafrost regions</title>
      <link>https://trid.trb.org/View/2709851</link>
      <description><![CDATA[The cooling performance of Duct-Ventilated Embankments (DVE) primarily depends on cold-energy superposition and the three-dimensional penetration of cooling capacity, both of which are controlled by duct spacing. This study investigates the spatiotemporal evolution of the embankment heat budget and temperature field under duct spacings of 2, 3, and 5 m within a future warming-climate scenario using long-term continuous highway monitoring data and three-dimensional numerical simulations. The better-performing DVE configuration (D2) is then benchmarked against a crushed rock-based embankment (CRBE) control case. As duct spacing increases from 2 to 5 m, the embankment center shifts from a cumulative net heat-release state to a cumulative net heat-absorption state, while the artificial permafrost table undergoes pronounced downward migration; the 5 m configuration demonstrates a clear degradation tendency. DVE also mitigates the sunny-shady slope effect. Duct spacings of 2 and 3 m promote the formation and persistence of a continuous, closed low-temperature core at depth, limiting the deep lateral temperature difference to within 0.2 °C and reducing the risk of asymmetric thaw settlement. Compared to the CRBE control case, the better-performing DVE case produces a slightly smaller absolute cooling intensity but establishes a thermal regime that remains more stable between years and more spatially homogeneous. In addition, DVE decreases the risk of performance attenuation associated with pore-structure degradation, enhancing long-term service reliability.]]></description>
      <pubDate>Fri, 18 Sep 2026 11:38:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2709851</guid>
    </item>
    <item>
      <title>Automatic identification of slope anomalies in traffic speed deflectometer measurements via machine learning</title>
      <link>https://trid.trb.org/View/2768715</link>
      <description><![CDATA[The Traffic Speed Deflectometer (TSD) is a technology for obtaining bearing capacity data of roads in a non-destructive manner without disrupting the traffic flow. It measures the deflection velocities of the loaded surface using Doppler sensors positioned at different offsets of the rear axle dual tires. The deflection slopes are then derived by dividing the deflection velocity by the instantaneous vehicle speed. For different reasons, a measured slope curve may deviate from the expected form, constituting an anomaly that can impact further processing. Analytical method alone is not enough to detect all anomalies in a set of measurements. Additionally, manual inspection of each measurement is not practical. As such, this study evaluates various Machine Learning (ML) methods for anomaly detection, including Isolation Forest and One-Class Support Vector Machine. A set of 257,040 simulated slope curves were used as reference shapes to train the models. These simulations use pavement structures from 408 pavement sections across the United States and Canada. The methods are then tested on the measurement data. For evaluation, a representative number of positively and negatively classified curves are visually inspected. The ML methods successfully learn to identify positive and negative patterns, demonstrating their potential as a promising methodology for first-level consistency control of TSD data.]]></description>
      <pubDate>Wed, 16 Sep 2026 16:15:47 GMT</pubDate>
      <guid>https://trid.trb.org/View/2768715</guid>
    </item>
  </channel>
</rss>