<?xml version="1.0" encoding="utf-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>TRB Publications Index</title><link>http://pubsindex.trb.org/</link><atom:link href="http://pubsindex.trb.org/common/TRIS Suite/feeds/rss.aspx?cdatein=1year" rel="self" type="application/rss+xml" /><description></description><language>en-us</language><copyright>Copyright © 2015. 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>TRB Publications Index</title><url>http://pubsindex.trb.org/Images/PageHeader-wTitle.png</url><link>http://pubsindex.trb.org/</link></image><item><title>Emerging Hazards in Commercial Aviation—Report 3: The Human Contribution to Safety</title><link>http://pubsindex.trb.org/view/2778090</link><description><![CDATA[Technology, operational practices, regulation, and the professionalism of the aviation workforce have helped commercial aviation achieve an extraordinary level of safety. Pilots, air traffic controllers, mechanics, dispatchers, inspectors, certifiers, and others in the workforce are an integral, measurable part of the aviation safety system. These human contributions can be a more visible, measurable, and actionable part of aviation system management. Recommendations in this report focus on a workforce data foundation, transforming safety analysis, and strengthening safety culture and oversight. This report, TRB Special Report 361, was requested by Congress in order to help identify, monitor, understand, and address emerging aviation safety risks. This report marks the third in a series of reports to be issued within a span of 10 years by the National Academies' Committee on Emerging Trends in Aviation Safety. Modern operational data and advanced analytics create new opportunities to identify emerging workforce-related hazards, understand successful performance, and act before adverse effects appear in traditional safety indicators.]]></description><pubDate>Fri, 18 Sep 2026 13:46:15 GMT</pubDate><guid>http://pubsindex.trb.org/view/2778090</guid></item><item><title>An Improved Tension Evaluation Approach for Smart Strands</title><link>http://pubsindex.trb.org/view/2778111</link><description><![CDATA[Monitoring strand tension is crucial for bridge safety assessment and reinforcement design. Smart strands, composed of a straight core wire embedded with a fiber grating and six helical wires, have been developed for tension monitoring. In this study, an improved tension evaluation method based on smart strands was developed. First, the external load-induced strain of the core wire was obtained by applying a developed complete temperature model to temperature and wavelength measurements of the grating, which can fully deduce the temperature-induced strain. Subsequently, an analytical iterative model, accounting for the off-axis effect of the helical wires, was derived to convert the core wire strain into the corresponding strain in the helical wires. Finally, the total tension of the smart strand was calculated by summing the contributions from the core and helical wires based on their respective strains. The proposed method was validated using laboratory experiments, demonstrating that the complete temperature compensation model exhibits universality across statically determinate and indeterminate structures, achieving an error reduction exceeding 55.6% in statically indeterminate systems. Furthermore, the proposed tension evaluation method achieves estimation errors below 2.5% and reduces maximum tension errors by 56.6% compared with conventional methods that directly use the core wire strain as the average strain. The results indicate that traditional temperature compensation models for optical fiber sensors cannot fully eliminate temperature effects, mainly because of the neglect of temperature field effects rather than sensor limitations.]]></description><pubDate>Wed, 16 Sep 2026 16:15:41 GMT</pubDate><guid>http://pubsindex.trb.org/view/2778111</guid></item><item><title>Unveiling the Coupled Spatiotemporal–Network Resilience of Urban Taxi Systems under Recurrent Daily Peak-hour Disturbances</title><link>http://pubsindex.trb.org/view/2778109</link><description><![CDATA[Urban taxi systems are continuously exposed to recurrent peak-hour demand surges that impose sustained operational stress, while existing resilience studies have predominantly focused on rare extreme disruptions. This study develops a coupled spatiotemporal–network resilience framework to investigate how urban taxi systems respond to routine, high-frequency peak-hour disturbances. Based on two weeks of high-resolution taxi trip data collected in Zhuhai, eight representative daily disturbance events are identified through a data-driven detection approach capturing distinct commuting and leisure travel demand regimes. Spatiotemporal resilience and network resilience are assessed separately considering resistance and adaptability at both node and system levels. These dimensions are subsequently integrated through an entropy-weighted coupling method to reveal multidimensional resilience states. Results show pronounced scenario-dependent resilience patterns. Structured commuting peaks maintain moderate operational performance despite limited structural flexibility, whereas leisure-oriented disturbances exhibit higher network robustness and greater route diversity, facilitating adaptation despite more dispersed operational demands. For all identified events, resilience is determined not by total demand volume, but by spatiotemporal distribution of trips and the alignment between operational performance and network topology. The coupled framework further uncovers scale-dependent effects. Spatiotemporal resilience exhibits significant spatial variations across disturbance scenarios, whereas network resilience remains comparatively stable owing to persistent structural properties. High-demand nodes often exhibit stronger integrated resilience owing to structural centrality, while system-level resilience is governed by demand dispersion and flow organization. By explicitly linking spatiotemporal operations to network topology under recurrent daily disturbances, this study lays a diagnostic foundation for scenario-adaptive resilience management in routine urban transportation operations.]]></description><pubDate>Wed, 16 Sep 2026 16:15:41 GMT</pubDate><guid>http://pubsindex.trb.org/view/2778109</guid></item><item><title>Career-Stage-Specific Training Priorities for State Department of Transportation Workforce Development: An Integrated Regression and Importance–Performance Analysis</title><link>http://pubsindex.trb.org/view/2778108</link><description><![CDATA[Transportation agencies nationwide face a convergence of workforce aging, accelerating retirements, and rapid technological change that threatens institutional capacity. Despite decades of research documenting these challenges, most workforce studies treat employees as a homogeneous population or rely on descriptive analyses that cannot disentangle career-stage effects from demographic confounders such as age, gender, and education. This study addresses three gaps: the absence of covariate-adjusted career-stage analysis, the lack of stage-specific training diagnostics, and limited within-agency evidence on life-cycle workforce needs. We analyze survey data from 386 employees at the Indiana Department of Transportation using regression models with dummy-coded career stages and demographic controls, applying Benjamini–Hochberg false discovery rate correction across multiple outcomes. The regression results are then translated into actionable priorities through stage-specific importance–performance analysis (IPA). Our findings revealed a systematic mid-career vulnerability: mid-career employees consistently gave lower ratings of training effectiveness and organizational support than both early-career and senior employees, even after controlling for demographic factors. Senior employees reported high technical proficiency but faced emerging gaps in technology-related domains. Education and gender emerged as independent predictors with distinct patterns across proficiency and satisfaction outcomes. The IPA diagnostics indicated that training priorities shifted systematically across career stages, from foundational communication skills for early-career staff to regulatory knowledge for mid-career staff and software competencies for senior employees. These findings were synthesized into a life-cycle workforce framework that offers targeted strategies for each career phase, providing a replicable approach for transportation agencies seeking to align training investment with stage-specific workforce needs.]]></description><pubDate>Wed, 16 Sep 2026 16:15:41 GMT</pubDate><guid>http://pubsindex.trb.org/view/2778108</guid></item><item><title>Alkali Threshold and its Correlation with Alkali–Silica Reactivity of Coarse Aggregates</title><link>http://pubsindex.trb.org/view/2777817</link><description><![CDATA[This study explores, for the first time, the use of the novel AASHTO T 416 alkali threshold test (ATT) standard to systematically evaluate the alkali threshold (AT) of 156 coarse aggregates of diverse mineralogy from the northeastern region of the United States. Combining the AT results with the reactivity classifications obtained from AASHTO TP 144-21 (T-FAST) revealed a precise portrayal of the alkali–silica reaction (ASR) susceptibility of this sample population. T-FAST detected alkali–silica reactive phases in most of the aggregates used in the study, classifying them as slow or moderately reactive. Compared to the commonly used mortar or concrete-based accelerated methods, T-FAST stood out as the most sensitive test for detecting the alkali–silica susceptibility of aggregates. The T-FAST classification and AT values of aggregates were combined to design prescriptive approaches. This study presents an example illustrating how the information from T-FAST and ATT can be used to minimize the ASR risk of concrete in the field and optimize mitigation strategies.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777817</guid></item><item><title>Transit Access, but at What Cost? Affordability Outcomes Across California’s Transit-Oriented Development Typologies</title><link>http://pubsindex.trb.org/view/2777816</link><description><![CDATA[As California continues to expand its investment in high-quality transit infrastructure, the need to align transit access with housing affordability has become increasingly urgent. In this study, we develop a replicable framework for classifying and evaluating over 60,000 high-quality transit station areas based on built environment characteristics and affordability outcomes. Using a 1.5-mi network-based pedestrian buffer, we apply 𝘒-means clustering to categorize station areas into four typologies: transit-oriented development (TOD), transit-supportive development, limited TOD, and transit-adjacent development (TAD). We then assess affordability using housing and transportation cost burdens across income levels, availability of income-restricted units, and statistical models, including random forest and Pearson correlation analysis. Results reveal that only 14% of station areas meet the criteria for full TOD, while TADs, comprising over 25%, exhibit the highest cost burdens and the lowest levels of affordable housing. Strong inverse correlations between TOD scores and affordability burdens indicate that transit-oriented urban form is closely linked to household cost efficiency. This framework offers a valuable decision-support tool for aligning land use, housing, and transportation policies to target resources where transit access and affordability are most misaligned.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777816</guid></item><item><title>Effects of Upstream Signal-Induced Arrival Interruptions on Delays at a Downstream Signalized Intersection</title><link>http://pubsindex.trb.org/view/2777815</link><description><![CDATA[Delay is a critical measure when evaluating level of service under interrupted flow conditions and is widely used to assess operational improvements and optimize traffic signals. However, existing delay models often fail to capture the effects of arrival interruptions, which are prevalent at closely spaced intersections, and are typically limited to specific facility types. To address this limitation, this paper proposes a generalized analytical delay model that explicitly accounts for the effects of arrival interruptions caused by upstream signals on delays at downstream signals. The model consists of two components that maintain consistency between delay terms while reflecting both the duration and timing of arrival interruptions. The model was evaluated against the Highway Capacity Manual (HCM) method at internal approaches of a typical diamond interchange and closely spaced arterial intersections. While the HCM model produced reasonable estimates under good progression conditions, it tended to overestimate delays under poor progression. This overestimation was attributed to double-counting of the effects of arrival interruptions; the variation in arrival rates was reflected through the progression factor, while capacity was simultaneously reduced. In contrast, the proposed model avoids this type of redundancy, demonstrating improved accuracy. Results show that it provided more reliable delay estimates across both facility types and various offset scenarios, with errors ranging from −4.6 to 1.9 s compared with VISSIM simulations or field measurements. The model is expected to support more robust delay estimations and informed decision-making under diverse traffic conditions and network configurations.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777815</guid></item><item><title>Comparison of Road Segment-Level Bicycle Volumes Estimation Models: Traditional Regression, Machine Learning, Large Language Models, and Novel Spatial Regression Approaches</title><link>http://pubsindex.trb.org/view/2777812</link><description><![CDATA[Estimating road segment-level bicycle volumes is essential for exposure-based safety analysis and infrastructure planning for vulnerable road users. Existing approaches often require extensive data collection, including short-term counts expanded using continuous counts or direct-demand models based on roadway, socioeconomic, and land-use characteristics. Recent methods incorporating machine learning, deep learning, and crowdsourced data have shown promise, but are often limited by data bias and inconsistent availability across jurisdictions. This study evaluates several benchmark methods, in addition to newly proposed approaches based on spatial proximity effects, including linear interpolation, Bayesian spatial negative binomial, random forest, large language models, graph convolutional networks (GCNs), and spatial lag regression (SLR). A dataset consisting of 1,127 road segments with a full year of bicycle count data (2025), along with site characteristics from the cities of Waterloo and Kitchener, Canada, was used to compute seasonal average daily bicycle volumes. Model performance was assessed using root mean squared error and mean absolute error across varying levels of data availability. The results indicated that the proposed SLR model consistently outperformed all other models across all data availability levels, including the GCN model. These findings demonstrate that incorporating spatial proximity and roadway characteristics can improve estimation accuracy without requiring large training datasets or complex model structures, providing a practical and interpretable solution for bicycle volume estimation in data-constrained environments.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777812</guid></item><item><title>Long-Term Field Performance Evaluation of Chip Seals</title><link>http://pubsindex.trb.org/view/2777783</link><description><![CDATA[The optimal timing and long-term effectiveness of chip seal treatments on newly constructed pavements remain uncertain, particularly for local agency roads. This study was an evaluation of the field performance of chip seals applied at different postconstruction intervals (from immediately to after 4 years) across three local agencies in Minnesota: Cass County, Crow Wing County, and the City of St. Cloud. Performance was monitored from 2015 to 2025 using the International Roughness Index (IRI), manual crack counts, and surface ratings. Linear mixed-effects models were used to assess performance trends. Results show that early application, either immediately after construction or within 1 year, was associated with better surface condition. However, binder type and reclaimed asphalt pavement content in the original pavement mix design exerted greater influence on long-term durability than chip seal timing alone. While chip seals were associated with preservation benefits, baseline material design largely dictated performance outcomes. Cracking data and IRI alone were insufficient for optimizing chip seal treatment schedules. The integration of material, structural, functional, and environmental factors may improve treatment timing decisions and maximize preservation benefits.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777783</guid></item><item><title>Toward Automatic Pavement-Condition Index Estimation: An Enhanced Multi-Sensor Deep Learning Framework for Comprehensive Pavement Distress Detection and Severity Assessment</title><link>http://pubsindex.trb.org/view/2777782</link><description><![CDATA[Timely and accurate pavement distress detection is essential for sustainable road maintenance. Traditional manual methods are costly, time-consuming, and prone to subjectivity. This study proposes a multi-sensor framework that integrates (Red-Green-Blue (RGB) imagery with low-cost depth sensing for automated detection, classification, and severity quantification of pavement distresses. The system identifies and categorizes 11 common distress types using convolutional neural networks and a sensor fusion strategy. Detected regions are projected onto three-dimensional point clouds to enable class-specific severity assessment. Pothole and rutting severity classification are examined as case studies. Using a low-cost Red Green Blue plus Depth (RGB-D) multi-sensor fusion framework, we report, for the first time, automated rutting severity classification integrated into Pavement-Condition Index (PCI) estimation as a proof-of-concept, alongside enhanced pothole severity quantification and classification. Results show high accuracy, with pothole severity correctly estimated in 96.2% of cases and rutting severity classified with 100% accuracy. The system also reliably detects shallow potholes as small as 15 mm in depth. By automating key measurements required for PCI estimation, this work lays a practical foundation for data-driven pavement evaluation. The proposed approach offers a scalable and cost-effective solution for infrastructure monitoring, reducing reliance on manual inspection and enhancing roadway management, while also demonstrating extensibility to other distress types.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777782</guid></item><item><title>Evaluating Transfer-Learning Strategies for Multi-Label Classification of Concrete Bridge Defects with Limited Data</title><link>http://pubsindex.trb.org/view/2777781</link><description><![CDATA[Bridge inspection images often contain multiple co-occurring surface defects within the same scene, yet much of the existing literature has emphasized single-label classification tasks. This study examines how transfer learning strategy affects multi-label classification of concrete bridge defects using the Concrete Defect Bridge Image (CODEBRIM) dataset and three ImageNet-pretrained convolutional neural network backbones: VGG16, ResNet50, and EfficientNetB0. Each backbone was evaluated under an add-on configuration with a frozen backbone and a selective fine-tuning configuration, both using a common multi-label prediction head with six sigmoid outputs corresponding to the defect classes and the Background class. To further examine the effect of backbone adaptation, performance sensitivity to fine-tuning depth was analyzed by progressively varying the number of trainable final backbone layers. The results showed that selective fine-tuning consistently outperformed the frozen-backbone add-on configuration across all three backbones, and that performance did not improve simply by unfreezing more layers. Instead, moderate selective fine-tuning provided the strongest overall results. Among the evaluated configurations, VGG16 with the last three backbone layers unfrozen achieved the best holdout test-set performance, with a subset accuracy of 0.840 and an F1-score of 0.874. These findings indicate that transfer-learning design choices, particularly the extent of backbone adaptation, play a central role in multi-label bridge-defect classification on limited engineering datasets.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777781</guid></item><item><title>An Observational Evaluation of Pedestrian Behavior at Signalized Intersections</title><link>http://pubsindex.trb.org/view/2777780</link><description><![CDATA[Pedestrians’ appropriate use of pedestrian infrastructure at signalized intersections—in particular, their compliance with the pedestrian signal and pushbutton use—can have important safety and efficiency implications. Therefore, it is important to understand why pedestrians do not always use infrastructure as engineers intend. However, there is a lack of research on pedestrian pushbutton use and pedestrian signal compliance, particularly at US locations. This research addresses this knowledge gap with an observational study of eight corners across four signalized intersections in Corvallis, Oregon, USA, where video data was collected and manually reduced to yield 2,633 unique pedestrian observations. At locations where pushbutton actuation was required to yield the WALK signal, over 80% of pedestrians traveling alone and arriving first at the corner pressed a pushbutton. For pedestrians who arrived on the DON’T WALK interval, results from a logistic regression model suggest that presence of conflicting vehicles on arrival, pedestrian waiting time, whether the pedestrian was first to arrive at the intersection corner on a cycle, and pedestrian pushbutton presence/use were significant factors in predicting the likelihood of the pedestrian entering on the DON’T WALK interval. These findings will help transportation practitioners better understand why—and the extent to which—pedestrians engage in “non-compliant” behaviors at signalized intersections and how engineering design choices can encourage compliance.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777780</guid></item><item><title>Utilizing ADS-B and Computer Vision for Runway Status Lights</title><link>http://pubsindex.trb.org/view/2777778</link><description><![CDATA[A technology-based runway incursion prevention system applicable to non-towered and general aviation airports and enabled by automatic dependent surveillance–broadcast (ADS-B) and computer vision is presented. The proposed system was designed to be a broadly applicable solution to prevent runway incursion accidents. A proof-of-concept surveillance prototype was developed and tested at the Purdue University Airport using simple and inexpensive hardware. The prototype was evaluated by comparing human observations with live and recorded operations for ADS-B and computer vision validation, respectively. Utilizing ADS-B, 94% of all operations observed during the study, including all operations involving ADS-B-transmitting aircraft, were detected and able to sufficiently provide timely runway status information despite being in airspace where ADS-B is not required. Meanwhile, using object detection and filtering algorithms, computer vision software designed to run on solar-powered modules was able to detect 110 out of 110 approaching aircraft while providing sufficient time to indicate potential traffic conflicts. Furthermore, the computer vision software was able to detect more than 94% of surface operations correctly. This study provides evidence that such an approach to runway surveillance can be effective. The proposed cost-effective surveillance methodology can provide numerous benefits to airport safety and operations.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777778</guid></item><item><title>Hot-Mix Asphalt Cracking Potential Prediction from Binder Rheological Parameters</title><link>http://pubsindex.trb.org/view/2777779</link><description><![CDATA[This study provides insights into the complexities of asphalt binder modifications and their effect on hot-mix asphalt (HMA) cracking potential. Binder rheological parameters were used to predict HMA cracking potential, in particular, the flexibility index (FI) of the Illinois Flexibility Index Test. An experimental program was developed to assess HMA cracking potential from binder characteristics. Various binder types, including binders modified with polymer softener, were considered. Rheological tests were conducted on binders at short- and long-term aging conditions. The Illinois FI test was conducted on mixes produced with the tested binders. Binder rheology parameters and corresponding FI results were analyzed statistically. Multilinear regressions were used to develop predictive models linking binder rheological parameters (independent variables) to HMA FI (dependent variable). The HMA FI was successfully predicted using binder and mix parameters available to contractors to allow binder selection. The binder rheology parameters are creep stiffness at 60 s (S) and rate of stress relaxation at 60 s (m), and the mix characteristics are binder content, asphalt binder replacement, and number of design gyrations (Ngyr). The models were developed for short- and long-term aged HMA. Three uncertainty categories for predicted FI from binder rheology parameters are introduced. The categories are associated with binder selection for acceptance based on potential cracking risk.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777779</guid></item><item><title>Predicting the Frequency and Duration of Turn-Bay Spillovers and Mutual Lane Blockages over Congested Arterials for Proactive Coordinated Signal Controls</title><link>http://pubsindex.trb.org/view/2777777</link><description><![CDATA[Accurate prediction of queue sizes, spillovers, and lane blockages is critical for urban signal control and traffic management but remains challenging due to the time-varying and spatially interdependent nature of traffic dynamics. These challenges are especially pronounced in congested commuting arterials, where upstream disruptions rapidly propagate and affect traffic states and queue formation at multiple downstream intersections. To contend with such challenges in design of proactive real-time traffic control, this study presents a multibranch, multihead long-short-term memory (LSTM) system for queue dynamic prediction, including queue distance, onset time, and duration of bay spillovers or lane blockages. By assigning each upstream intersection its own dedicated LSTM branch, the proposed prediction model can learn and preserve location-specific temporal patterns at the target location and then exert its multihead fusion layer to receive and integrate these features to capture the information of cumulative traffic states and their impacts from upstream intersections. The proposed LSTM system is structured to reflect the spatial relations between a target intersection and its upstream intersections, and to maintain a balance between spatial resolution and computational efficiency. Performance evaluations using extensive simulations over MD 355 in Bethesda, Maryland, show that the proposed system achieves a 100% detection rate for all queue blockages, with exact predictions of onset times and durations and no false alarms. Compared with benchmark models such as the extended Kalman filter and standard recurrent neural networks, the proposed system demonstrates strong robustness in capturing spatial dependencies and persistent congestion patterns, key attributes for proactive real-time traffic signal control.]]></description><pubDate>Tue, 15 Sep 2026 09:48:14 GMT</pubDate><guid>http://pubsindex.trb.org/view/2777777</guid></item></channel></rss>