<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>Evolution of isolated heaving defects for vertical rail displacements: a multi-disciplinary study</title>
      <link>https://trid.trb.org/View/2655762</link>
      <description><![CDATA[Track heaving defects are likely caused by multiple complex factors, including irregular superstructure displacement and frost-induced expansion of subgrades and subsoils. These defects can significantly disrupt normal train operations. Early detection and trend prediction of such defects are essential for implementing effective predictive maintenance, enabling maintenance prioritization, and reducing both costs and operational disruptions. However, rail geometry profiles often exhibit non-stationary and complex characteristics, making the identification of isolated defects challenging. While most current studies focus on evolution of defect magnitude, the impact of defect length on train–track interaction remains largely unexplored. Furthermore, understanding the natural degradation process of isolated defects is complicated by maintenance activities that disrupt the development of track defects. To address these challenges, a multi-disciplinary approach was employed, incorporating signal processing techniques, multibody dynamics simulations, and geotechnical site investigations to identify, characterize, and analyse the effects of defect evolution on the train-track system. A 50-km section of measurement train data, collected over a five-year period, was analysed to examine the impact of both defect amplitude and length evolution on train performance, including operational risks and passenger comfort. Statistical analysis of the location, magnitude, and length of isolated heaving defects, along with their evolution patterns, was conducted. The findings were synthesized into an evolution diagram of train performance based on changes in defect profiles and applied to real-world railway data.]]></description>
      <pubDate>Thu, 09 Apr 2026 10:08:55 GMT</pubDate>
      <guid>https://trid.trb.org/View/2655762</guid>
    </item>
    <item>
      <title>Axle load violations model for sustainable financing of road pavement maintenance in Nigeria</title>
      <link>https://trid.trb.org/View/2684430</link>
      <description><![CDATA[The sustainability of Nigeria’s federal highway network is increasingly undermined by persistent axle load violations among heavy goods vehicles (HGVs). Overloaded axles accelerate pavement deterioration, escalate lifecycle maintenance costs, and compromise freight system reliability. This study integrates engineering-based deterioration modelling with operations management principles to estimate pavement damage costs and develop a penalty-based financing framework. Using weigh-in-motion (WIM) systems, traffic and axle load data were collected across three freight-intensive corridors, namely Lokoja-Abuja, Ilorin-Jebba, and Abakaliki-Ogoja. Equivalent Single Axle Load (ESAL) analysis and econometric modelling were employed to quantify incremental damage and calibrate penalty functions. Findings reveal systemic overloading, with corridor-specific damage costs ranging from ₦0.74 to ₦5.70 per ESAL and violation rates exceeding 70% on high-intensity routes. A log-linear penalty model was developed, explaining over 80% of the variability in cost recovery estimates. The study demonstrates that monetizing axle load violations through calibrated penalties can transform enforcement into a sustainable financing mechanism. The contribution lies in extending operations management theory by embedding asset management, externality internalization, and game-theoretic principles into road infrastructure governance. The proposed model offers a scalable framework for enhancing infrastructure resilience, optimizing maintenance funding, and improving regulatory compliance in Nigeria and other Sub-Saharan African economies.]]></description>
      <pubDate>Wed, 08 Apr 2026 13:41:51 GMT</pubDate>
      <guid>https://trid.trb.org/View/2684430</guid>
    </item>
    <item>
      <title>A study on fast recognition of multi-class bridge damage based on improved Siamese ResNet50 with limited samples</title>
      <link>https://trid.trb.org/View/2635538</link>
      <description><![CDATA[To solve the issue of poor accuracy and low computational efficiency of multi-class structural damage recognition based on Siamese Neural Network (SNN) with limited training samples, improved Siamese ResNet50 is proposed in this paper, in which the Siamese frameworks are set to run separately by reprogramming. More specifically, the side of the network of the labeled images is frozen while the other side are continuously iterated separately. The structural appearance images dataset including health pavement, four types of cracks, steel corrosion and concrete spalling was constructed and used to train and test the improved Siamese ResNet50. The results were compared with CNN (VGG16, MobileNetV3-Large, ResNet50) and the corresponding Siamese Network (Siamese VGG16, Siamese MobileNetV3-Large) using the same dataset.The results indicate that the classification speed of the improved Siamese ResNet50 is improved from 1 image/second to 11.24 image/second, with an increased rate of 1024 %. The average Overall Accuracy (OA) and Macro-F1 are 94.25 % and 93.87 %, respectively. For samples with challenging situations, the improved Siamese ResNet50 also provides a satisfactory accuracy of 88.4 %. In the case of limited samples, the results of the Axiom-based Gradient-weighted Class Activation Mapping (XGrad-CAM) show that the CNNs cannot extract features correctly even the satisfactory accuracy is achieved. The results of the t-Distributed Stochastic Neighbour Embedding (t-SNE) show that the improved Siamese ResNet50 can achieve effective multi-class structural damage recognition by Contrastive Loss.]]></description>
      <pubDate>Thu, 02 Apr 2026 16:58:39 GMT</pubDate>
      <guid>https://trid.trb.org/View/2635538</guid>
    </item>
    <item>
      <title>A Comparison Between Ground Penetrating Radar (GPR) and Impact Echo (IE) for Detection of Common Concrete Bridge Decks Defects</title>
      <link>https://trid.trb.org/View/2165291</link>
      <description><![CDATA[The condition of the transportation network in the United States is deteriorating and will require enormous efforts to significantly improve its state. Out of the 600 thousand bridges, almost 14% is structurally obsolete and another 14% is functionally obsolete. Concrete bridge condition assessment is a challenge to bridge inspectors because of the variety of problems that may exist. Therefore, the need for innovative inspection and maintenance programs is rising and new technologies are being applied to facilitate the detection of different defects that affect concrete bridges. An important component of the inspection and rehabilitation of concrete bridges is the ability to assess the condition of the bridge's deck. The advent of nondestructive evaluation techniques aided this task. Several nondestructive techniques have been successfully utilized to detect common defects in concrete bridge decks. This paper will study Ground Penetrating Radar (GPR) and Impact Echo (IE) abilities in the detection of the most common concrete bridge deck defects faced by inspectors using fabricated bridge decks with simulated defects. These simulated decks will be used to validate and compare the abilities of both methods in detecting the presence of cracks, delaminations, and voids.]]></description>
      <pubDate>Sun, 29 Mar 2026 17:20:02 GMT</pubDate>
      <guid>https://trid.trb.org/View/2165291</guid>
    </item>
    <item>
      <title>A Novel Mode Sharp-Based Damage Detection Method of Large-Span Bridges</title>
      <link>https://trid.trb.org/View/2164372</link>
      <description><![CDATA[Health monitoring of large-span bridges mainly aims at continuously accumulated damages and thus damage detection is the core technique of bridge health monitoring systems. Any crack or localized damage on a structure reduces its stiffness, which will lead to reduction in the natural frequency and modification of the vibration modes of the structure. The damage detection is based on comparison of initial signatures (frequency, mode shapes and so on) of intact bridge with those of damaged bridge. To identify the damage in time, the sensibility to damage detection is the key factor. In this paper, a new damage index, W value, is proposed by comprehensively utilizing the existing indexes. According to the analogue analysis the damages of Wenhui cable-stayed bridge with finite element method, considering 3% noise which in order to account the measurement error, the index, W value could localize the damage precisely and its validity is verified. The numerical results show that the proposed index is convenient to be calculated and has highly sensible. This method can be used in a practical bridge health monitoring system to realize the damage identification early.]]></description>
      <pubDate>Sun, 29 Mar 2026 17:20:02 GMT</pubDate>
      <guid>https://trid.trb.org/View/2164372</guid>
    </item>
    <item>
      <title>A memory-augmented multimodal network with physics-based feature fusion for defect detection in shipbuilding</title>
      <link>https://trid.trb.org/View/2685329</link>
      <description><![CDATA[Automated defect detection is critical for ensuring structural safety in shipbuilding industrial systems. Traditional image-based detection methods face challenges in diverse defects, especially for large-scale components with complex geometries in marine engineering. Although multimodal image-point cloud fusion offers enhanced capability, current approaches face ineffective multi-view point cloud fusion, insufficient semantic feature extraction, and significant information loss during cross-modal integration. To address these challenges, this study proposes a novel multimodal framework for efficient and accurate defect identification in large-scale ship components. A pre-fusion mechanism based on physics-space-based texture mapping for high-resolution point cloud generation from multi-view scans is proposed for multimodal inspection. Furthermore, an improved point cloud feature extractor emphasizing center prediction is designed to capture rich semantics. Additionally, a memory-augmented multimodal fusion detection module is developed to preserve the original image, geometric, and fused cross-modal features, thereby mitigating information loss. The proposed method surpasses state-of-the-art algorithms in surface defect detection, achieving a 0.939 P-AUROC on in-situ workpieces and demonstrating superior detection accuracy. This study establishes an effective paradigm for automated quality inspection in safety-critical marine engineering applications, providing foundational support for preventing critical operational failures.]]></description>
      <pubDate>Fri, 27 Mar 2026 10:14:49 GMT</pubDate>
      <guid>https://trid.trb.org/View/2685329</guid>
    </item>
    <item>
      <title>Interpretable machine learning classification for vehicle response-based railway track irregularities assessment</title>
      <link>https://trid.trb.org/View/2656976</link>
      <description><![CDATA[Diligent railway track irregularities maintenance scheduling is important for a safe rail traffic operation. Machine learning classifiers based on vehicle response have been shown to be an effective approach for track irregularities assessment. However, when simultaneously assessing several types of track irregularities within a given track section, loss of interpretation becomes a prevalent issue as the track irregularities assessment and the accompanying machine learning classification become more complicated. The present work examines the use of machine learning classification for combined track irregularities assessment and the subsequent result interpretation using Shapley Additive Explanation (SHAP) and Accumulated Local Effects (ALE). Testing results of the trained classification models show a high accuracy value, i.e., higher than 92%, with a low sensitivity against change in operational parameters. This indicates the suitability of this technique for track irregularities assessment. Furthermore, the interpretation analyses demonstrate a favourable potential in interpreting the outcomes of the track irregularities classification for specific sections. In particular, information from SHAP and ALE can be useful for identifying the threshold for acceptable acceleration levels. This feature is especially valuable for exploring the root causes of irregularities within a given track section. The interpretability not only enhances the ability to diagnose and address specific track irregularities but also underscores the potential for data-driven and data-informed approaches in the domain of railway track assessment.]]></description>
      <pubDate>Fri, 27 Mar 2026 10:13:31 GMT</pubDate>
      <guid>https://trid.trb.org/View/2656976</guid>
    </item>
    <item>
      <title>M-CLUSTER: multistage clustering for unsupervised train wheel condition monitoring</title>
      <link>https://trid.trb.org/View/2647001</link>
      <description><![CDATA[Wheel out-of-roundness (OOR) is a common wheel defect that raises maintenance costs and increases the risk of failure or damage to track components. This paper proposes a novel multistage clustering framework (M-CLUSTER) for unsupervised condition monitoring of train wheels. The framework initially extracts time-domain features from raw acceleration responses collected by rail-mounted sensors. Sensitive features are then selected through an unsupervised feature selection algorithm called local learning-based clustering (LLC). Next, a detector model is trained using density-based spatial clustering of applications with noise (DBSCAN), a data clustering method effective for clusters with similar density. Since this algorithm does not originally involve separate training and testing phases, a new two-step mechanism is introduced: (1) training on a healthy dataset and (2) testing on an unlabelled dataset. Finally, the severity of train wheel defects is classified by K-means, with cluster validity indices (CVI) automatically determining the number of severity clusters (classes). The framework’s efficiency is demonstrated through the detection of defective wheels using the Alfa Pendular passenger model. Results indicate that M-CLUSTER accurately identifies train wheel flats and polygonal wear without labelled data, achieving 98% accuracy by selecting 10 features from the set.]]></description>
      <pubDate>Wed, 25 Mar 2026 16:41:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2647001</guid>
    </item>
    <item>
      <title>Effect of operating conditions on the time-frequency characteristic of high-speed vehicle-turnout dynamic interaction using synchrosqueezed wavelet transform</title>
      <link>https://trid.trb.org/View/2647000</link>
      <description><![CDATA[Structural irregularities are a primary source of vibration when vehicles traverse turnouts. This study investigates their impact on the time-frequency characteristics of vehicle dynamic responses under varying operating conditions, focusing on the crossing zone of a No. 18 ballastless high-speed railway turnout. A vehicle-turnout coupled dynamic model incorporating rail flexibility is developed, and the time-frequency features of vehicle responses under different conditions are analysed. The results reveal that structural irregularities in the crossing zone lead to higher peak values and more pronounced high-frequency components (250–850 Hz) in vertical wheel/rail forces in the facing direction compared to the trailing direction. Furthermore, the vertical vibration acceleration of the point rail in the facing direction exhibits higher peak values, greater energy in dominant frequency components, and the emergence of a new dominant frequency band (70–150 Hz). Differences in vertical wheel/rail forces and point rail vertical vibration acceleration across vehicle types are primarily evident in the high-frequency range (above 150 Hz). Additionally, increasing speed amplifies dynamic response amplitudes, peak values, and vibration frequencies. This study provides critical insights into the dynamic response characteristics of turnouts under various conditions, offering a foundation for damage detection and operational assessment.]]></description>
      <pubDate>Wed, 25 Mar 2026 16:41:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2647000</guid>
    </item>
    <item>
      <title>A Long-Term Experimental Study on Evaluating Corrosion Currents in Reinforced Concrete for Marine Structures</title>
      <link>https://trid.trb.org/View/2676062</link>
      <description><![CDATA[This study investigates the corrosion currents in four reinforced concrete mixes—SL (cement replacement of 50% slag), FA (cement replacement of 20% fly ash), T1 (cement replacement of 50% slag and 20% fly ash), and T2 (cement replacement of 20% fly ash and 8% silica fume)—using an electromigration method to accelerate chloride transport and initiate corrosion within weeks/months. The corrosion propagation was monitored over 1600 days using electrochemical techniques such as Linear Polarization Resistance (LPR), Electrochemical Impedance Spectroscopy (EIS), and Galvanostatic Pulse (GP) measurements. The study evaluated how the composition of the concrete mixes and reservoir length influenced corrosion, by testing both binary mixes (SL and FA) and ternary mixes (T1 and T2). The results show that the LPR readings, which uses prolonged polarization, generally produces higher corrosion current values than GP readings, offering a dynamic view of corrosion but greater variability. The binary mixes with slag (SL) or fly ash (FA) exhibit higher corrosion currents, while ternary mixes, especially those containing fly ash and silica fume (T2), show reduced corrosion currents, suggesting improved resistance. The larger reservoir length contributed to higher corrosion currents, highlighting the critical influence of exposure conditions, concrete mix composition, and measurement techniques in evaluating corrosion. This underscores the importance of considering these factors collectively when assessing the durability and long-term performance of concrete in corrosive environments.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:57 GMT</pubDate>
      <guid>https://trid.trb.org/View/2676062</guid>
    </item>
    <item>
      <title>Multisource road defect dataset featuring brick pavement and multi-perspective views for intelligent detection</title>
      <link>https://trid.trb.org/View/2681286</link>
      <description><![CDATA[Timely and automated detection of road surface defects is essential for ensuring pavement safety and optimizing maintenance operations. This paper presents Multisource Road Defect Dataset (MRDD), a high-quality dataset designed to support the development of automated pavement inspection systems. MRDD covers asphalt, concrete, and brick pavements and includes six common defect types. A hybrid image acquisition strategy combining UAVs and smartphone imagery enhances collection efficiency and viewpoint diversity, closely aligning with real-world applications. A total of 35 object detection models, including YOLO series and Faster R-CNN, were benchmarked on MRDD, with YOLOv9e achieving 97.0% mean Average Precision. Results demonstrate the dataset's adaptability across diverse detection architectures and its value in supporting stable, high-precision defect recognition. By aligning with the sensing and data requirements of intelligent inspection systems, MRDD serves as both a reliable training source and a practical benchmark for advancing visual perception, defect analysis, and decision-making in infrastructure automation.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:49 GMT</pubDate>
      <guid>https://trid.trb.org/View/2681286</guid>
    </item>
    <item>
      <title>Influence of the geometric parameters of the connecting pipeline on the stiffness and damping of the pneumatic spring suspension at high-speed rolling stock</title>
      <link>https://trid.trb.org/View/2646998</link>
      <description><![CDATA[The study presents investigation is the pneumatic spring suspension system of diesel train DPKr-3. The research aims to estimate the influence of track irregularities on the pneumatic spring dynamic stiffness and its damping coefficient at 170–250 km/h speeds. The pneumatic spring suspension system characteristics were studied using an improved mathematical model of the system ‘rolling stock-track’. The pneumatic spring ‘force-strain’ relations are obtained at different values of the connecting pipeline diameter and length at high speeds. It was determined that increasing a connecting pipeline diameter from 20 to 40 mm and a speed change from 170 to 250 km/h causes decreasing a spring dynamic stiffness. It is shown that at a speed range from 170 to 250 km/h and a connecting pipeline diameter from 20 to 40 mm, a change in a pipeline length from 3 to 6 m does not lead to significant changes in stiffness. It was found that in the range of speeds from 170 to 250 km/h, the damping coefficient has maximum values with a diameter of the connecting pipeline from 25 to 30 mm.]]></description>
      <pubDate>Tue, 24 Mar 2026 09:10:16 GMT</pubDate>
      <guid>https://trid.trb.org/View/2646998</guid>
    </item>
    <item>
      <title>Damage diagnosis method for girder-hinge joint systems in multi-girder bridges using field-identified influence surfaces</title>
      <link>https://trid.trb.org/View/2670790</link>
      <description><![CDATA[Stiffness degradation of hinge joints and main girders is common damage in multigirder bridge systems. If these damages are not detected and reinforced in time, the load transverse transmission ability of a multigirder bridge will be reduced, which will lead to unbalanced load distribution and the failure of the girders. To solve this problem, this paper proposes a joint diagnosis method of girder and hinge joints damage of multigirder bridges based on moving load test and field identified influence surface. Firstly, an applicable influence surface identification method is developed using the time-spatial coincidence approach of vehicle location and bridge responses. Then, based on hinge-jointed plate method, the relationship between influence surface, hinge joints shear stiffness and girder bending stiffness is established. Two damage indexes: the hinge joint damage index based on field-tested versus theoretical influence surface differences, and the girder stiffness degradation index derived from their integral ratio, are constructed. Finally, the proposed method is verified through a prefabricated hollow slab beam bridge moving load testing and damage identification example. The proposed method can effectively identify typical damages of multigirder bridges in the early stage, which can guide the reinforcement and management of such bridges.]]></description>
      <pubDate>Mon, 23 Mar 2026 15:25:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2670790</guid>
    </item>
    <item>
      <title>Effects of Pavement Distresses on Cyclist Behavior: Observations in Metro Manila</title>
      <link>https://trid.trb.org/View/2669853</link>
      <description><![CDATA[Road defects are a common problem in the Philippines. Their presence on bicycle lanes affects the behavior of cyclists, putting their safety at risk. This study analyzed the behavior of cyclists towards road defects on Class II, unprotected, and Class III, shared, bicycle lanes in Quezon City. The presence and severity of road defects along bicycle lanes were surveyed and Cyclists’ Behavior Survey was conducted in ten locations. Selected behavior of cyclists towards pavement distresses were observed, including evasion, swerving, lane change, passing through the distress, and speed change. The road defects considered in this study are delamination, potholes, scaling, raveling, and depression. The results showed evasion as the most common behavior of cyclists when approaching road defects. It also reveals that the cyclists' behavior varies based on the types and severity of road defects based on their perceived risk on their safety.]]></description>
      <pubDate>Mon, 23 Mar 2026 15:21:00 GMT</pubDate>
      <guid>https://trid.trb.org/View/2669853</guid>
    </item>
    <item>
      <title>Prediction for freeze–thaw cycles induced degradation of dynamic shear modulus in Qinghai loess: structural damage concept-based model-Case study</title>
      <link>https://trid.trb.org/View/2647604</link>
      <description><![CDATA[Freeze-thaw cycles (FTCs) and earthquakes are key factors contributing to foundation deformation and instability in loess engineering in cold regions. Seismic loading acting on loess weakened by FTCs can lead to catastrophic failure(e.g., landslides, settlement). FTCs degrade loess stiffness by causing microstructural damage. To quantify this damage and explore its impact on dynamic properties, this study examined the influence of FTCs on the microstructure and dynamic property of loess from Qinghai Province, China using scanning electron microscopy (SEM) and cyclic simple shear (CSS) tests. Image-Pro Plus(IPP) was used to quantitatively analyze changes in pore and particle structures. The results show that after FTCs, soil particles break apart, become more angular, and rearrange. The total pore area increases, with a higher proportion of large and medium pores. The dynamic shear modulus exhibits a total attenuation of 64.64%, with the first cycle contributing up to one-third of the overall reduction. A random forest model identified key microscopic characteristic parameters governing the degradation of the initial dynamic shear modulus. According to damage theory and considering the evolution of pore and particle structures, structural parameters representing loess integrity were proposed. A dynamic shear modulus prediction model was then developed, which effectively accounts for structural disturbance. These findings advance geohazard forecasting by coupling microstructural damage metrics with dynamic stiffness, offering a novel tool for slope-stability and subgrade-settlement assessments in cold region loess.]]></description>
      <pubDate>Fri, 20 Mar 2026 14:47:52 GMT</pubDate>
      <guid>https://trid.trb.org/View/2647604</guid>
    </item>
  </channel>
</rss>