Search Results (333)

Sustainable and resilient road infrastructure requires the early identification of pavement deterioration mechanisms that emerge under complex urban traffic conditions, particularly at signalized intersections where repeated stop–go operations, queue persistence, and lane-wise freight concentration generate highly nonuniform structural loading. However, most existing intelligent transportation studies emphasize crash prediction, traffic-state estimation, or mobility optimization, while the infrastructure-performance consequences of freight-dominant interrupted flow remain insufficiently addressed. To support proactive pavement management and resilient urban road operation, this study proposes a traffic simulation-driven deep learning framework for predicting lane-level pavement deterioration under freight loading and stop–go urban traffic conditions. A high-resolution PTV Vissim 2024 microscopic simulation environment was developed for a four-leg signalized urban intersection, and a structured multi-scenario design was used to generate progressively increasing operational stress regimes, ranging from baseline flow to freight-dominant oversaturated operation. The resulting lane-wise dataset integrates direct traffic variables with pavement-oriented descriptors, including the Lane Freight Loading Index (LFLI), Stop–Go Severity Index (SGSI), ESAL proxy, queue persistence, and Loading Asymmetry Index (LAI). To learn the complex relationship between traffic operation and infrastructure degradation, a new Freight-Aware Lane Interaction Transformer Network (FLIT-Net) is introduced. The proposed model combines feature embedding, lane-interaction self-attention, freight-aware gating, residual refinement, and multi-task regression to jointly predict rutting risk, fatigue-cracking risk, and the Pavement Deterioration Index (PDI). Experimental results show that FLIT-Net outperforms MLP, CNN, LSTM, Bi-LSTM, and generic Transformer baselines, achieving RMSE/MAE/$R^2$ values of 0.041/0.032/0.9687 for rutting risk, 0.044/0.034/0.9635 for fatigue-cracking risk, and 0.031/0.024/0.9824 for PDI. Sensitivity and scenario-wise analyses further confirm that deterioration increases monotonically with freight intensity, stop–go severity, and queue persistence, highlighting the importance of lane-resolved deterioration intelligence for sustainable maintenance prioritization. The proposed framework bridges traffic microsimulation, pavement-oriented feature engineering, and freight-aware deep learning, providing a decision-support basis for improving the performance, safety, and resilience of urban pavement infrastructure. Full article

Pedestrian safety at signalized intersections is a key component of sustainable urban mobility, as safer walking environments support active transportation, reduce crash risk, and improve the inclusiveness of urban transport systems. This study presents a structured review of pedestrian–vehicle conflicts based on a systematic PRISMA-guided literature search, synthesizing 60 studies with emphasis on operational conditions, behavioral factors, infrastructural characteristics, and surrogate safety measures. The review examines the application of surrogate safety measures (SSMs), including Time-to-Collision (TTC), Post-Encroachment Time (PET), Pedestrian Path Deviation (PPD), and Deceleration-to-Safety Time (DST). The findings reveal significant variability in threshold definitions and methodological approaches, which limits the comparability and transferability of results across different traffic contexts. Building on this synthesis, the paper proposes an integrated conceptual framework linking behavioral, operational, and infrastructural determinants to conflict occurrence and severity. The analysis shows that existing studies often treat these factors in isolation, reducing the generalizability of current models. Overall, this review identifies key methodological inconsistencies in surrogate safety indicators and emphasizes the need for standardized yet context-sensitive thresholds and locally validated conflict models to improve the comparability and transferability of pedestrian–vehicle conflict assessments. Full article

Crash severity prediction is critical for emergency response, infrastructure spending, and risk communication. Although machine learning has been widely applied to this problem, three gaps prevent practical deployment: uncalibrated probability scores, SHAP-based explanations whose faithfulness has not been verified, and models never tested outside their training jurisdiction. The proposed framework, SAE-XCrash (Safety-Aware and Explainable Crash Severity Prediction), addresses all three using two public datasets—US-Accidents (7.0 million records, 2016–2023) and UK STATS19 (approximately 1,010,000 records, 2016–2022)—with strict temporal splits throughout. Notably, the US-Accidents severity label measures traffic disruption duration, not injury outcome; results should be interpreted accordingly. Previously unknown label-schema drift led to a revised binary target with Severity 4 as the only positive class. Five classifiers are compared. Post hoc isotonic calibration reduces Expected Calibration Error by 97.3% at negligible discrimination cost. A four-step quantitative SHAP audit confirms statistically significant deletion faithfulness; however, explanation stability fails at realistic perturbation levels (54.3% low-stability fraction at sigma = 0.05), driven by spatial data sparsity in sparse geohash cells—a negative result that carries direct operational implications for deployment. A three-tier cross-dataset transfer experiment (zero-shot, recalibration, full retrain) shows that temporal features transfer robustly across jurisdictions, while spatial memorization is the primary generalization barrier. All code, split indices, and model artifacts are publicly available. Full article

Background: Mountain biking is increasingly popular but carries a large risk of severe trunk and spinal injuries. However, realistic crash scenarios for back protector design remain poorly characterized. This study aimed to define trunk impact conditions during mountain biking crashes. Methods: A multi-body model for mountain bike accident reconstruction was developed, and its kinematics were validated against real-world crash video footage. The model was then used to assess the influence of initial conditions (speed, slope, crash cause, etc.) on trunk impact kinematics (velocities, forces, pseudo-energy) and spinal loading indicators during forward crashes. Results: Across 288 simulated crashes, the median normal trunk impact velocity (4.61 m/s) and pseudo-energy (48 J) aligned with current test standards, while substantial tangential (5.97 m/s) and rotational (4.90 rad/s) components were also observed. Three main impact types emerged: head–thorax impacts (43.5%), involving a head impact followed by chest impact (Vn: 5.42 m/s, Emax: 59 J); tumbling (25.1%), featuring a head impact followed by back impact (Vn: 3.98 m/s, Emax: 57 J); and overflip–back impacts (20.7%), involving direct back contact (Vn: 3.35 m/s, Emax: 47 J). Conclusion: This study’s results define trunk impact conditions during MTB crashes, informing on realistic boundary conditions for testing and designing back protectors. Full article

Urban traffic accidents emerge from complex interactions among traffic instability, roadway structure, environmental disturbance, and temporal dynamics, yet many existing prediction approaches still treat accident risk as a discrete classification problem over isolated observations. This study proposes a Continuous Hazard Field Intelligence Framework for Traffic Accident Emergence and Urban Safety Prediction, which models hidden urban danger as a topology-aware spatio-temporal hazard field that evolves continuously across connected transportation infrastructure. The framework integrates heterogeneous urban traffic observations, including incident records, crash data, roadway attributes, temporal cues, and contextual risk factors, into a unified hazard-aware learning pipeline. A dedicated preprocessing strategy combines topology-constrained spatial alignment, temporal hazard window embedding, risk-diffusion feature lifting, hazard-sensitive normalization, and continuous hazard surface initialization to convert fragmented event-centered observations into a smooth and learning-ready hazard representation. A structured deep learning architecture is then developed to perform spatial hazard encoding, temporal hazard evolution, continuous hazard reconstruction, and localized accident emergence prediction. Experimental evaluation was conducted on two large-scale real-world traffic safety datasets, namely the XTraffic Incident Dataset (2022–2024) with 1,441,904 records and the Motor Vehicle Collisions–Crashes Dataset with 2,026,647 records. All model configurations were evaluated under the same experimental setting, using the same dataset-specific preprocessing protocol, a 70/30 train–test split, and identical evaluation metrics. The final CHFI configuration achieves 99.12% accuracy, 98.94% precision, 98.76% recall, 98.85% F1-score, and 0.998 AUC on Dataset 1, and 98.63% accuracy, 98.41% precision, 98.16% recall, 98.28% F1-score, and 0.997 AUC on Dataset 2. Compared with the initial non-hazard-aware baseline configuration evaluated under the same data split and evaluation protocol, the final CHFI model improves the F1-score by 7.91 percentage points on Dataset 1 and 8.26 percentage points on Dataset 2. These results indicate that the proposed hazard-field formulation can improve accident-emergence prediction within the controlled experimental setting, while the reported gains should be interpreted relative to the specified baseline and evaluation design. Full article

Road intersection crashes remain a major contributor to injuries due to complex conflict patterns and multimodal interactions. Among the factors influencing intersection safety, inadequate intersection sight distance (ISD) attributed to roadside sight obstructions can limit drivers’ ability to respond to conflicting movements, potentially affecting crash injury outcomes. Despite its importance, visual obstruction has rarely been examined as a distinct context in traffic crash injury severity modeling. This study investigates crash injury severity at visually obstructed intersections using an ordinal probit modeling framework applied to 951 intersection crashes documented with sight obstruction as a contributing factor in Wyoming over the period 2014 through 2023. Crash data were analyzed to identify the effects of driver behavior, vehicle characteristics, roadway geometry, environmental conditions, and traffic control on ordered injury severity outcomes ranging from property damage only (PDO) to fatal and serious injury. Nonparametric bootstrap resampling with 1000 iterations was employed to assess parameter stability and construct empirical confidence intervals. Average marginal effects were estimated to quantify the change in probability of each injury severity level associated with key predictors. The results indicate that alcohol involvement produces the largest severity shift, reducing the probability of PDO outcomes by 51.2 percentage points while increasing the probability of fatal and serious injury by 34.2 percentage points. Hillcrest grade locations increase fatal and serious injury risk by 14.4 percentage points, while adverse road surface conditions, including snowy, icy, and wet pavements, consistently reduce fatal and serious injury probability by 12.5 to 15.1 percentage points, reflecting behavioral adaptation to visually salient hazard cues. Bootstrap validation confirms strong parameter stability across all estimates, with 94% of parameters showing bootstrap standard errors within 25% of their asymptotic counterparts. By formally establishing visually obstructed intersections as a dedicated severity modeling context and integrating systematic bootstrap validation, this study contributes both substantive and methodological insights to support evidence-based prioritization of intersection safety improvements. Full article

Real-time crash risk assessment is a key component of proactive road safety management, enabling the identification of hazardous conditions within short temporal intervals before crashes occur. Traditional crash-based models are unsuitable for such applications due to the rarity, reporting delay, and stochastic nature of crash data. Traffic conflicts, capturing near-miss interactions between road users, provide a practical alternative for real-time safety analysis. Over the past decade, numerous modelling approaches have been developed to translate conflict information into crash risk estimates; however, the literature remains fragmented and lacks a unified analytical synthesis. This review presents a state-of-the-art, model-centric analysis of conflict-based approaches, classifying them into five paradigms: statistical/regression-based, Bayesian, extreme value theory (EVT), machine learning (ML), and hybrid models. Beyond classification, the study conducts a structured cross-paradigm comparison across key dimensions, including conflict representation, data characteristics, temporal modelling, uncertainty treatment, validation strategies, computational complexity, and operational readiness. The paradigms are further interpreted through the complementary lenses of conflict frequency and severity. The review identifies key research gaps, including fragmented conflict definitions, challenges in modelling rare and extreme events, incomplete treatment of uncertainty and spatiotemporal dynamics, and limitations in validation, transferability, and deployment. Emerging research directions include standardized and adaptive conflict indicators, EVT–machine learning integration, integrated uncertainty-aware frameworks, advanced spatiotemporal modelling, transferable models, and scalable real-time implementation. By combining structured evidence mapping and cross-paradigm synthesis, this study supports model selection, development, and deployment for dynamic crash risk assessment. Full article

Advanced Driver Assistance Systems (ADAS) and Automated Driving Systems (ADS) are becoming convenient modes of transportation; however, their safety remains a critical concern as crashes continue to occur. To reveal crash trends and temporal variations, this study develops time-series forecasting models to predict future crash counts of such vehicles. The crash dataset released by the National Highway Traffic Safety Administration (NHTSA) has been used here. Two univariate forecasting models—the Seasonal Autoregressive Integrated Moving Average (SARIMA) and the Facebook Prophet model—have been used here for different datasets. The models were trained on 30 months of data (July 2021 to December 2023) and validated on 6 months of data (January–June 2024). Validation metrics include Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Theil’s U1 statistic. Results showed that Facebook Prophet significantly outperformed SARIMA for both datasets, achieving an RMSE of 2.71 and an MAPE of 6.9% for ADAS, and an RMSE of 2.24 and an MAPE of 8.85% for ADS. For both systems, the model revealed empirically observed cyclical patterns and consistent rising trends. ADAS crashes exhibit a bimodal temporal pattern, with recurring peaks in January and May–June, alongside notable troughs in February–March and August–September. ADS displays a trimodal pattern, with recurring peaks in April–May, August and October, alongside notable troughs in December and the early winter months. These patterns represent empirically identified temporal regularities rather than causally attributed seasonality. From the future forecasts for July to December 2024, the model showed that ADAS crashes are expected to range between 40 and 80 per month, while ADS crashes are projected to remain between 20 and 40 per month. These findings underscore the need for proactive safety measures and enhanced regulatory oversight during identified high-risk periods to mitigate the growing trend in AV crashes. Full article

Crashes involving freight and commercial vehicles impose substantial human and economic costs, yet most severity studies pool vehicle types or focus exclusively on heavy trucks, masking class-specific risk mechanisms. This study estimates separate Random Parameters Binary Logit models with heterogeneity in means and variances for three vehicle categories—heavy-duty multi-axle trucks (n = 6512), two-axle trucks (n = 2656), and light-duty pickup trucks (n = 23,477)—using 32,645 crash records from Thailand’s national highway network (May 2022–December 2024). Pairwise transferability tests rejected parameter transferability, with four of six comparisons exceeding the 97 percent confidence level (three of these above 99 percent; χ² = 85.38 to 240.01), confirming that disaggregate estimation is statistically warranted. Three core findings emerge: First, although barrier medians, cut-in-front maneuvers, and sideswipe crashes affect severity in consistent directions across all vehicle types, their magnitudes differ sharply: the protective effect of barrier medians is nearly six times larger for two-axle trucks (ME = −0.160) compared to heavy-duty trucks (ME = −0.028). Second, several determinants are class-specific: dark unlit conditions elevate severity only for two-axle trucks (ME = 0.128), flush medians only for heavy-duty trucks (ME = 0.040), and raised medians only for light-duty pickups (ME = 0.042). Third, no random parameter is common to all three models. Pooled models, therefore, impose misleading homogeneity assumptions; vehicle-type-specific estimation is essential for targeted safety policy. Full article

During freeway reconstruction and expansion, median crossover sections where traffic is maintained during construction are vulnerable to changes in lane configuration, abrupt geometric changes, and construction interference. These factors may lead to safety risks and operational efficiency losses. Existing studies have mainly relied on microscopic traffic simulation to evaluate speed limit schemes, while engineering costs, environmental impacts, driver responses, and policy constraints have rarely been considered in an integrated manner. This study proposes a two-stage evaluation framework that integrates VISSIM microscopic traffic simulation, the Entropy Weight Method–Technique for Order Preference by Similarity to an Ideal Solution (EWM–TOPSIS), and the Fuzzy Analytic Hierarchy Process (FAHP). A four to eight-lane freeway expansion project in a plain area of northern China is used as the case study. Field speed data from a representative median crossover section are used for model calibration and speed-pattern analysis. A total of 27 simulation scenarios is then constructed by combining three bottleneck types, three traffic saturation levels, and three speed limit schemes. The EWM–TOPSIS results show that the 80→70 km/h scheme achieves the highest relative closeness in all scenarios. The FAHP evaluation, based on six criteria and 21 indicators, also ranks this scheme first. Its ranking remains unchanged under ±10% criteria weight perturbations. Field speed comparison indicates that vehicles exhibit a deceleration–recovery pattern when passing through the crossover opening. Overall, the 80→70 km/h gradual speed reduction scheme can be regarded as a candidate scheme for work zones with similar median crossover configurations. Under localized calibration conditions, it can provide decision-making support for reducing operational disturbances, fuel consumption, and external impacts associated with crash risk. Full article

Accurately predicting traffic crash severity is critical for intelligent transportation systems, where outcomes emerge from the interaction of infrastructure, environment, traffic control, and human behavior. However, existing approaches face three key challenges: severe class imbalance, computational inefficiency, and limited support for system-level risk understanding. To address these issues, this study proposes a unified and system-aware framework integrating Conditional Tabular Generative Adversarial Network (CTGAN), Tabular Prior-data Fitted Network (TabPFN), and eXplainable Artificial Intelligence (XAI) methods for data augmentation, efficient prediction, and interpretable analysis. CTGAN enhances rare but critical crash states while preserving feature dependencies; TabPFN enables accurate multi-class prediction with limited dataset-specific tuning; and XAI methods quantify the influence of key factors and their interactions. Experiments on a real-world crash dataset from Boston show that the proposed framework achieves competitive predictive performance with less reliance on dataset-specific hyperparameter tuning, while also providing complementary interpretability results from multiple perspectives. The results further reveal that crash severity is jointly shaped by visibility, traffic control, roadside features, and temporal dynamics, highlighting the interconnected nature of risk within the transportation system. By integrating predictive modeling with complementary interpretability analysis, the framework provides a systems-oriented basis for examining how environmental, infrastructural, and temporal conditions jointly relate to crash severity in the studied urban crash data, while offering a methodological reference for broader safety applications that require further validation. Full article

This paper provides a systematic review of recent applications of NLP methods for analyzing traffic crash reports, with a focus on estimating crash severity, crash duration, and crash causation. The review covers prior research using probabilistic topic modeling methods such as LDA, STM, and hierarchical Dirichlet processes in addition to research using transformer-based language models, which include encoder-based models like BERT and PubMedBERT as well as decoder-based models like GPT, GPT2, ChatGPT, GPT-3, and LLaMA. The review starts with a systematic literature selection process with predefined inclusion criteria. We categorize the reviewed studies into the following application areas: crash severity prediction, risk factor identification in crashes, and road safety analysis. The results show several complementary advantages of using different NLP techniques to achieve different analytical goals. Topic models allow for interpretable and exploratory pattern discovery, while encoder models are well-suited for structured prediction problems. Decoder models have the additional flexibility to perform zero-shot and few-shot reasoning, which makes them useful for reasoning about under-sampled or under-reported data. Across the literature, hybrid methods that combine text and structured data outperform individual methods in terms of prediction accuracy and broad applicability. Challenges across the literature include class imbalance, lack of standardization in preprocessing and evaluation methods, and the tradeoff between prediction accuracy and interpretability of prediction models. These findings highlight the importance of aligning model selection with data availability and operational constraints, pointing toward future research directions in hybrid modeling frameworks, standardized evaluation protocols, and real-world deployment of NLP-driven traffic safety systems. Full article

Accurately estimating the severity of crash injuries among older drivers is paramount for enhancing traffic safety, a task challenged by class imbalance and label noise. Traditional predictive paradigms often struggle to identify rare severe cases, as they tend to prioritize global accuracy, thereby compromising sensitivity to high-risk outcomes. To overcome these limitations, this study develops a Log-Loss Cleaned and Probability-Calibrated Cascade (L-CSC) framework by strategically integrating existing advanced algorithmic components for robust and reliable severity prediction. Initially, a Log-Loss-based noise filtering mechanism is implemented to purge outliers and ambiguous samples from the training data, thereby enabling higher-quality representation learning. Subsequently, a two-stage cascade architecture is designed to decouple the classification task. Stage I employs a Preliminary Screening Model, optimized via Bayesian optimization for F2-score, to specifically maximize the recall for severe and fatal cases. In Stage II, a Stacking ensemble classifier is deployed to achieve a fine-grained classification of injury levels among the cases identified in the initial screening. Finally, Isotonic Regression is employed to calibrate the output probabilities from both stages, ensuring that the resulting risk estimations are statistically sound and reliable. Empirical evaluations demonstrate that the L-CSC framework effectively balances overall performance with critical risk detection, achieving a robust Macro-F1 of 0.7296. Specifically, compared to the best-performing baseline, the recall and F1-score for the critical severe and fatal category showed relative improvements of over 82% and 62%, respectively. Ablation analyses further substantiate the vital contributions of both the data cleaning and calibration modules. This research demonstrates that the cascaded framework effectively mitigates the biases inherent in imbalanced datasets, providing a robust algorithmic foundation to potentially support future traffic safety interventions. Full article

Effective incident management is a cornerstone of transportation system performance, influencing roadway clearance times (RCTs) and the risk of secondary crashes. This study investigated how towing regulations involving large commercial vehicle crashes and jurisdictional variations affect the management of large-vehicle crashes, focusing on the relationship between regulatory frameworks, incident duration, and secondary crash occurrence with the state of Tennessee as a case study. The objective was to determine whether differences in towing policies, operational mandates, and rural/urban contexts lead to measurable changes in clearance efficiency. A multi-year dataset of more than 770,000 traffic incidents and 4400 towing-involved large-vehicle crashes from 2017 to 2022 was analyzed. Statistical methods, including two-sample testing and hazard-based survival modeling, were applied to evaluate the impact of towing regulations and operational protocols on roadway clearance and secondary crash patterns. The results consistently showed that strong performance-based towing regulations, such as mandated maximum response times and standardized training and equipment requirements, were associated with significantly lower average RCTs. Jurisdictions with enforced rapid-response mandates achieved average clearance durations of approximately 120–130 min, even under high incident volumes, compared to over 150 min in areas without performance benchmarks or with more complex procedural requirements. A pronounced rural–urban divide was observed, with incidents outside urbanized areas averaging 30–40% longer clearance times, largely due to limited towing resources, longer dispatch distances, and less stringent regulatory enforcement. Secondary crash analysis identified that more than 90% of secondary collisions were linked to crashes requiring towing, with the majority occurring within 20 min and 0.5 miles of the primary incident, underscoring the direct connection between delayed clearance and safety risk. These results carry direct implications for transportation policy and incident management practice by providing empirical evidence that standardized, performance-based towing regulations can meaningfully reduce RCTs and secondary crash risk, particularly when paired with investments in rural towing infrastructure Full article

This study investigated injury severity in 18,528 police-reported vehicle-to-pedestrian crashes involving elderly pedestrians in legally classified local areas of South Korea during 2012–2021. Injury severity was coded into four ordered categories: fatal, serious, minor, and reported injury. To stabilize scenario extraction from a categorical crash database, an integrated screening workflow was applied, including near-zero-variance filtering, redundancy control among overlapping roadway encodings, representative-variable selection within redundant groups, and chi-square association checks. Classification and regression tree (CART) modeling was then used to identify rule-based combinations of environmental, roadway, driver, pedestrian, and vehicle factors associated with elevated severity, while tree complexity was controlled through cost-complexity pruning and 10-fold cross-validation. A scenario-based sensitivity analysis was further conducted to evaluate counterfactual shifts in severity distributions under targeted control of key conditions within representative high-risk scenarios. The results showed that severe outcomes were concentrated in stacked-risk combinations rather than in single factors alone. A dominant pathway involved nighttime conditions combined with maneuver-related driving contexts and speeding-related violations. High-fatality scenarios persisted even when speed-related predictors were excluded, underscoring the roles of nighttime exposure, visibility limitations, conflict-prone roadway settings, heavy-vehicle involvement, and pedestrian exposure behaviors. The proposed framework translates administrative crash records into concise, operationally interpretable scenarios and intervention-relevant evidence for local-area safety. Full article