Journal of Advanced Transportation最新文献

To alleviate expressway congestion caused by excessive private vehicle use, trip reservation has emerged as a proactive traffic management strategy. However, when too many vehicles are admitted within the same time window, the travel efficiency of reservation users deteriorates, compromising the strategy’s effectiveness. Conversely, admitting too few vehicles leads to underutilization of road resources and degrades the operational performance of adjacent roads. This study addresses this challenge by identifying the optimal reservation quota. A reservation-based travel strategy is proposed for key corridors in urban road networks, comprising expressway segments and their parallel surface streets. The initial quota is determined through a dual-threshold bottleneck breakdown analysis, which estimates the capacity of reservation segments under varying service levels. A bilevel programming model is subsequently developed to allocate traffic flow across the network based on the initial quota. Simulation results reveal that the reservation quota significantly affects the performance of the network. The optimal quota lies between 70% of the theoretical maximum capacity and the prebreakdown threshold, within which the key corridor network maintains moderate traffic conditions. Compared to the no-reservation scenario, the average travel speed of reservation vehicles more than doubles (from 25.86 km/h to above 52.12 km/h), while the average travel delay is reduced by over 77% (from 774.77 s to below 179.01 s). The service level of reservation segments improves to Level C. Moreover, the strategy imposes minimal adverse effects on parallel surface streets, where average speeds decrease by less than 31% but remain above 22 km/h. These findings validate the effectiveness of the key corridors trip reservation system and confirm the optimal reservation quota range.

Preventing severe injuries in crashes has emerged as a central concern in freeway traffic safety research. To mitigate severe injuries, it is essential that the influential factors affecting accident severity be identified. In this research, accident data were collected from Los Angeles County, California, USA, freeways in the years 2016–2019, aggregating five influencing factors from five perspectives, including temporal factors, environmental factors, accident factors, accident participant factors, and traffic factors. A copula Bayesian network modeling approach was developed which combines a Bayesian network with a copula function to depict the interrelationships among crash severity outcomes and various influencing factors. The approach has the following advantages: (1) It has a more reasonable and interpretable structure. (2) It makes up for the limitation of traditional Bayesian networks that can only analyze discrete features by enabling the handling of both discrete and continuous variables. The copula Bayesian network reasoning analysis further demonstrates that various interconnections exist among different factors, and that accident type, lighting conditions, alcohol involvement, and average occupancy are the most critical contributors to fatal or severe injury accidents.

The accident rate of interchange ramps based on investigated Chinese cases is approximately two times higher than that of mainline sections, where losses in human lives and economic costs caused by heavy-duty truck (HDT) accidents are far greater than those of sedans. Nevertheless, existing risk assessments overlook the coupled effects of human–vehicle–road–environment factors, primarily focusing on the single-directional driving risk assessment of HDT longitudinal braking or lateral skidding. This study proposes a visual assessment method to evaluate the comprehensive lateral and longitudinal driving risk of HDTs on interchange ramps, utilizing floating vehicle data that incorporate the coupling effects of multiple factors. Based on 800 floating vehicle data samples of HDTs from 11 types of ramps, this study integrates driver experience, moderate adverse environments, and lateral/longitudinal acceleration distribution into the G–G diagram (longitudinal acceleration plotted versus lateral acceleration) to define safety thresholds. A mathematical model was fitted in Table Curve 2D to establish the basis for proposing the Driving Risk Index (DRI) and driving risk grading (DRG). Furthermore, precise geospatial matching and visualization of driving risks are achieved using a geographic information system (GIS). The method is validated from multiple dimensions, including statistical methods, surrogate safety measures, and comparison with existing models. Both empirical and statistical analyses demonstrated a strong correlation between the visualized distribution of DRI and route alignment. Moreover, validated by the coefficient of variation (CV), the model achieves an accuracy rate of 85.9%, exhibiting 28.2% and 15.5% higher performance than the two groups of comparative methods, respectively. This integrated approach from data processing to visualization overcomes traditional limitations and supports ramp optimization and intelligent early-warning systems.

Electric vehicles (EVs) are pivotal for reducing transportation-related emissions; however, the lack of adequate charging infrastructure remains a significant barrier to their widespread adoption. This study presents a comprehensive methodology for optimizing EV charging station placement. It combines a gravity model, scenario analysis, and mixed-integer linear programming (MILP) to ensure a thorough and robust approach. The model aims to maximize accessibility by ensuring both path-level and overall system demand coverage across diverse scenarios, providing reassurance about the validity of the findings. The methodology is tested on the Bursa–İzmir motorway in Turkey, a strategic intercity route with rapidly growing EV penetration. Results reveal that the optimal configuration involves locating charging stations in seven of the nine service areas. This allocation secures a minimum path coverage ratio of 0.903, meaning 90.3% of the route is covered by charging stations, and an overall demand coverage ratio of 0.935, indicating that 93.5% of total demand is covered across all scenarios. A sensitivity analysis further shows that increasing the network to 45 chargers elevates reachability levels to above 97%, indicating the infrastructure scale required for reliable service quality. The findings underscore the practical applicability of the proposed framework, providing policymakers and infrastructure planners with robust, data-driven guidance for charging network expansion. By integrating demand forecasting with resilient optimization, this study advances both methodological and empirical insights, empowering the audience to make informed decisions for sustainable EV adoption.

The growing demand for truck freight in the United States has intensified the shortage of truck parking, posing safety and operational challenges. While real-time Truck Parking Information and Management Systems (TPIMSs) offer current availability, predictive insights remain limited. This study develops hybrid machine learning and deep learning models to forecast truck parking utilization for both pretrip and en-route decision-making. A site-specific gradient boosting model achieved the best pretrip performance (average root mean square error [RMSE] = 0.154), while a long short–term memory–based truck parking site utilization prediction (TPSUP) model provided accurate en-route predictions (RMSE = 0.0429) with a one-hour horizon. To enhance usability, a “Popular Times” panel was designed to visualize predictions through intuitive, color-coded charts. These tools support safer and more efficient parking decisions, laying the groundwork for a more robust and predictive TPIMS.

Efficient timing of aircraft docking and push-back operations is crucial for enhancing the efficiency and reliability of civil aviation operations. Traditional methods suffer from significant data loss, high human involvement, and low accuracy, which are prone to inaccuracies that can disrupt airport scheduling and resource allocation. This paper introduces a reliable computer vision approach, YUAR (YOLOv7-UAVMOT Aircraft Recognition), which leverages advanced detection algorithms and machine learning to improve accuracy and reduce human error in monitoring aircraft movements. Utilizing a newly developed Image and Video Dataset of Aircraft on Airport Surface (IV-AAS), YUAR incorporates the YOLOv7-based Aircraft Detection (YAD) algorithm with UAVMOT for dynamic tracking. This integration facilitates a multithreshold frame interpolation method, significantly enhancing the precision of tracking aircraft docking and push-back events. Experiment results show that the system achieves a mean average precision (mAP) of 94.8% and an IDF1 score of 92.7%, demonstrating superior performance compared to existing methods such as YOLOv5 and DeepSORT by reducing identification switches. Additionally, the recognition rate of the docking and push-back times under various operational scenarios reaches 100% with minute-level precision. Our research offers significant implications for Airport Collaborative Decision Making (A-CDM), optimizing the allocation of resources and improving the overall operational efficiency of airports.

Escalating urban traffic problems are impeding city development, underscoring the critical need to better coordinate built environments (BEs) with traffic system states (TSSs). However, the efficiency measurements in the current data envelopment analysis (DEA) model exhibit excessive dependence on input and output data. The determination of weight constraints is also based on subjective judgment. This causes the varying impacts of critical and noncritical indicators on interaction dynamics to be ignored. This study introduces an enhanced DEA model with spatially adaptive weights calibrated by geographically weighted regression (GWR). We propose a TSS indicator that integrates three critical dimensions: traffic efficiency, traffic safety, and travel comfort. In this study, Jinan City is selected as the research area. The coordination assessments refined by incorporating constraints reveal significant disparities: 7.66% are fully coordinated and 63.7% show coordinated conditions, while 28.61% exhibit limited or no coordination. Compared to conventional DEA, the GWR–DEA model demonstrates marginally improved performance, validating the effectiveness of optimized weighting constraints in spatial coordination analysis.

The increasing scale of highway reconstruction and expansion projects has intensified traffic management challenges in construction zones, particularly within sparse road networks constrained by limited diversion capacities and elevated freight truck ratios. This study proposes an integrated analytical framework that combines microscopic simulation, automated machine learning (AutoML), and explainable artificial intelligence. Traffic flow dynamics under high truck proportions (72%) were modeled using the VISSIM microsimulation, generating 1320 parameterized scenarios encompassing traffic volume, work zone length, speed limits, and vehicle composition. By leveraging the AutoGluon AutoML framework, we developed an ensemble delay prediction model using optimized feature engineering. SHapley Additive exPlanations (SHAP) interpretability analysis further decoded the multifactorial coupling mechanisms influencing traffic organization. The results demonstrate that while complex ensembles achieved the lowest error (RMSE = 1.49), the CatBoost_BAG_L1 model was identified as the optimal model for operational deployment, achieving identical accuracy with a more than 25-fold improvement in computational speed. The SHAP-based interpretation revealed traffic volume as the dominant delay contributor, exhibiting nonlinear dynamics with escalating marginal effects beyond 1400 vehicles/h. Increasing the speed limit to 80 km/h elevated delays by 0.58 units, while work zones exceeding 2 km in length induced length-proportional delay amplification. This methodology advances intelligent decision-making for dynamic lane control and truck scheduling optimization in diversion-constrained environments.

The purpose of this study was to explore the risk factors for autonomous vehicle (AV) crashes and their interdependencies. A total of 659 AV crash data were collected between 2018 and July 2024 from AV crash reports published by the California Department of Motor Vehicles. Characteristics such as crash location and time, driving patterns, vehicle motion, crash type and vehicle damage, and traffic conditions were considered as potential risk factors in the study. Considering the multilevel and multidimensional nature of the crash data, the study adopted an association rule mining (ARM) approach to identify the risk factors that frequently occur in AV crashes. By improving the Apriori algorithm, based on the traditional Apriori algorithm, the association rule judgment index is added, and the accuracy and mining efficiency of association rules are improved. The results show that rear-end collisions in autonomous driving mode are more serious, especially when stopping at intersections, while the rear vehicle chooses to continue driving or slow down. Accident risk is higher at night, with on-street parking and 2-lane conditions in both directions. The occurrence of no-damage and minor crashes is more likely to be influenced by roadway characteristics and traffic conditions, and nonmotorized lanes, on-street parking, and median strips on the roadway play a key role in reducing crash damage. The results of the study inform relevant policies to improve road safety and the efficiency of AVs to enhance the overall safety of road traffic.

Driver deviations from planned routes during navigation threaten road safety and reduce traffic efficiency, with F-shaped intersections emerging as a high-risk scenario. This study investigates deviation causes using real-world navigation operation data and hourly aggregated observations. A generalized structural equation model (SEM) with a zero-inflated negative binomial link is applied to disentangle direct effects of external conditions and indirect effects mediated by traffic flow and congestion. Key findings include the following: exit/entrance roads (β = 0.135, p < 0.001) and road type (β = 0.100, p < 0.001) exert the strongest direct effects on deviations, while traffic congestion mediates 12% of the indirect effect of weather conditions on deviations. An actionable design takeaway is that extending advance signage on high-risk segments reduces deviations by 18% [−14%, −22%] at median traffic flow, providing targeted technical support for F-shaped intersection optimization to improve road safety and traffic efficiency.