Search Results (1,685)

Long-term multivariate time series forecasting is crucial for real-world applications, including energy consumption, traffic flow, healthcare, and finance. Usually, some statistical approaches are used for predicting future observations based on historical temporal data. Recently, transformer-based models with patch mechanisms have demonstrated significant potential in enhancing computational efficiency. However, their inability to fully capture intra-patch temporal dependencies often limits the accuracy of predictions. To address this issue, we propose the Frequency Compensation Patch-wise transFormer (FCP-Former), which integrates a frequency compensation layer into the patching mechanism. This layer extracts frequency-domain features via Fast Fourier Transform (FFT) and incorporates them into patched data, thereby enriching patch representations and mitigating intra-patch information loss. To verify the feasibility of this model, FCP-Former was conducted on eight benchmark datasets via PyTorch 2.4.0 and trained on an NVIDIA RTX 4090 GPU (Santa Clara, CA, USA). Results demonstrate that FCP-Former 48 optimal experiment results and 17 suboptimal experiment results across all datasets. Especially on the ETTh1 dataset, it achieves an average MSE of 0.437 and an average MAE of 0.430, while on the Electricity dataset, it achieves an average MSE of 0.186 and an average MAE of 0.277. This demonstrates that FCP-Former has better forecasting accuracy and a superior ability to capture periodic and trend patterns. Full article

Lane-changing maneuvers critically influence traffic flow and safety. This study introduces the Global Optimal Lane-Changing (GOLC) model, a framework that optimizes decisions by quantitatively predicting their systemic effects on surrounding traffic. Unlike traditional models that focus on immediate neighbors, the GOLC model integrates a kinematic wave model to precisely quantify the spatiotemporal impacts on the entire affected platoon, striking a balance between local vehicle actions and global traffic efficiency. Implemented in the Simulation of Urban Mobility (SUMO) environment, the GOLC model is evaluated against benchmark models Minimizing Overall Braking Induced by Lane Changes (MOBIL) and SUMO LC2013. Comparative evaluations demonstrate the GOLC model’s superior performance. In a three-lane scenario, the GOLC model significantly enhances traffic efficiency, reducing average delay by 3.4% to 46.8% compared to MOBIL under medium- to high-flow conditions. It also fosters a safer environment by reducing unnecessary lane changes by 1.1 times compared to the LC2013 model. In incident scenarios, the GOLC model shows greater adaptability, achieving higher average speeds and lower travel times while minimizing speed dispersion and deceleration. These findings validate the effectiveness of embedding macroscopic traffic theory into microscopic driving decisions. The model’s unique strength lies in its ability to predict and minimize the collective negative impact on all affected vehicles, representing a significant step towards real-world implementation in Advanced Driver-Assistance Systems (ADAS) and enhancing safety in next-generation intelligent transportation systems. Full article

We propose a Transformer-based data augmentation framework with a time-series dual-stream architecture to address performance degradation in encrypted network traffic classification caused by class imbalance between attack and benign traffic. The proposed framework independently processes the complete flow’s sequential packet information and statistical characteristics by extracting and normalizing a local channel (comprising packet size, inter-arrival time, and direction) and a set of six global flow-level statistical features. These are used to generate a fixed-length multivariate sequence and an auxiliary vector. The sequence and vector are then fed into an encoder-only Transformer that integrates learnable positional embeddings with a FiLM + context token-based injection mechanism, enabling complementary representation of sequential patterns and global statistical distributions. Large-scale experiments demonstrate that the proposed method reduces reconstruction RMSE and additional feature restoration MSE by over 50%, while improving accuracy, F1-Score, and AUC by 5–7%p compared to classification on the original imbalanced datasets. Furthermore, the augmentation process achieves practical levels of processing time and memory overhead. These results show that the proposed approach effectively mitigates class imbalance in encrypted traffic classification and offers a promising pathway to achieving more robust model generalization in real-world deployment scenarios. Full article

Urbanization and traffic congestion caused by the surge in car ownership have exacerbated energy consumption and carbon emissions, and dynamic traffic flow prediction and energy-saving route planning have become the key to solving this problem. Dynamic traffic flow prediction accurately captures the spatio-temporal changes of traffic flow through advanced algorithms and models, providing prospective information for traffic management and travel decision-making. Energy-saving route planning optimizes travel routes based on prediction results, reduces the time vehicles spend on congested road sections, thereby reducing fuel consumption and exhaust emissions. However, there are still many shortcomings in the current relevant research, and the existing research is mostly isolated and applies a single model, and there is a lack of systematic comparison of the adaptability, generalization ability and fusion potential of different models in various scenarios, and the advantages of heterogeneous graph neural networks in integrating multi-source heterogeneous data in traffic have not been brought into play. This paper systematically reviews the relevant global studies from 2020 to 2025, focuses on the integration path of dynamic traffic flow prediction methods and energy-saving route planning, and reveals the advantages of LSTM, graph neural network and other models in capturing spatiotemporal features by combing the application of statistical models, machine learning, deep learning and mixed methods in traffic forecasting, and comparing their performance with RMSE, MAPE and other indicators, and points out that the potential of heterogeneous graph neural networks in multi-source heterogeneous data integration has not been fully explored. Aiming at the problem of disconnection between traffic prediction and path planning, an integrated framework is constructed, and the real-time prediction results are integrated into path algorithms such as A* and Dijkstra through multi-objective cost functions to balance distance, time and energy consumption optimization. Finally, the challenges of data quality, algorithm efficiency, and multimodal adaptation are analyzed, and the development direction of standardized evaluation platform and open source toolkit is proposed, providing theoretical support and practical path for the sustainable development of intelligent transportation systems. Full article

The increasing shift towards battery electric vehicles (BEVs) in urban environments raises the question of how real-world traffic conditions affect their energy consumption. While BEVs are expected to reduce local emissions, their total energy demand, particularly in city traffic with with low average speeds, and therefore a higher impact of secondary consumption, remains insufficiently understood. To address this, a simulative framework to analyze the average energy consumption of an all-electric vehicle fleet in a mid-sized city, using Darmstadt, Germany, as a case study, is presented. A validated microscopic traffic simulation is built based on 2024 data and enriched with representative powertrain models for various vehicle classes, including passenger cars, trucks, and buses. The simulation allows the assessment of consumption under different traffic densities and speeds, revealing the substantial influence of secondary consumers and traffic flow on total energy demand. Furthermore, the study compares the $C{O}_2$ emissions of an all-BEV fleet with those of a fully combustion-based fleet. The findings aim to highlight the role of secondary consumers in urban traffic and to identify the potential for energy-saving. Full article

Quick User Datagram Protocol Internet Connection (QUIC) is a modern transport protocol leveraging the User Datagram Protocol (UDP) to improve latency, security, and mobility. In this study, we analyzed QUIC traffic by uploading a 10 MB file under varied maximum transmission unit (MTU), bandwidth, and segment size conditions. Interarrival times (IAT) at both client and server were captured and analyzed using ordering points to identify the clustering structure (OPTICS) clustering and Markov modelling. Transition matrices and eigenvalue spectra revealed steady states, convergence behavior, and spectral gaps. The results showed that parameter variations significantly affected the traffic state diversity and flow dynamics, optimizing QUIC performance in real-world deployments. Full article

Freeway weaving sections’ states under adverse weather exhibit characteristics of randomness, vulnerability, and abruption. A deep learning-based model is proposed for traffic state identification and prediction, which can be used to formulate proactive management strategies. According to traffic characteristics under adverse weather, a hybrid model combining Random Forest and an improved k-prototypes algorithm is established to redefine traffic states. Traffic state prediction is accomplished using the Weather Spatiotemporal Graph Convolution Network (WSTGCN) model. WSTGCN decomposes flows into spatiotemporal correlation and temporal variation features, which are learned using spectral graph convolutional networks (GCNs). A Time Squeeze-and-Excitation Network (TSENet) is constructed to extract the influence of weather by incorporating the weather feature matrix. The traffic states are then predicted using Gated Recurrent Unit (GRU). The proposed models were tested using data under rain, fog, and strong wind conditions from 201 weaving sections on China’s G5 and G55 freeway, and U.S. I-5 and I-80 freeway. The results indicated that the freeway weaving sections’ states under adverse weather can be classified into seven categories. Compared with other baseline models, WSTGCN achieved a 3.8–8.0% reduction in Root Mean Square Error, a 1.0–3.2% increase in Equilibrium Coefficient, and a 1.4–3.1% improvement in Accuracy Rate. Full article

Highway on-ramp merging areas represent critical bottlenecks that significantly impact traffic efficiency and sustainability. This paper proposes a novel Delay-Compensated Merging Control (DCMC) framework that addresses the practical challenges of cloud-based cooperative vehicle control under realistic communication conditions. The system integrates an efficient mixed-integer linear programming (MILP) model for trajectory optimization with a robust two-stage delay compensation mechanism. The MILP model coordinates mainline and ramp vehicles through proactive gap creation and speed harmonization, while the compensation framework addresses both deterministic and stochastic communication delays through Kalman filter-based prediction and real-time trajectory correction. Extensive simulations demonstrate that the DCMC system prevents traffic breakdown at near-capacity conditions (2200 vehicles per hour), achieving up to 31.6% delay reduction and 16.4% travel time improvement compared to conventional merging operations. The system maintains robust performance despite 2 s mean communication delays with 30 ms standard deviation, validating its readiness for practical deployment. By effectively balancing computational efficiency, safety requirements, and communication uncertainties, this research provides a viable pathway for implementing cloud-based cooperative control at highway merging bottlenecks to enhance both traffic flow efficiency and environmental sustainability. Full article

This study aims to guide the management and service of highways towards a more efficient and intelligent direction, and also provides intelligent and green data support for achieving sustainable development goals. The forecasting of traffic flow at highway stations serves as the cornerstone for spatiotemporal analysis and is vital for effective highway management and control. Despite considerable advancements in data-driven traffic flow prediction, the majority of existing models fail to differentiate between directions. Specifically, entrance flow prediction has applications in dynamic route guidance, disseminating real-time traffic conditions, and offering optimal entrance selection suggestions. Meanwhile, exit flow prediction is instrumental for congestion and accident alerts, as well as for road network optimization decisions. In light of these needs, this study introduces an enhanced heterogeneous spatiotemporal graph network model tailored for predicting highway station traffic flow. To accurately capture the dynamic impact of upstream toll stations on the target station’s flow, we devise an influence probability matrix. This matrix, in conjunction with the covariance matrix across toll stations, updated graph structure data, and integrated external weather conditions, allows the attention mechanism to assign varied combination weights to the target toll station from temporal, spatial, and external standpoints, thereby augmenting prediction accuracy. We undertook a case study utilizing traffic flow data from the Chengdu-Chengyu station on the Sichuan Highway to gauge the efficacy of our proposed model. The experimental outcomes indicate that our model surpasses other baseline models in performance metrics. This study provides valuable insights for highway management and control, as well as for reducing traffic congestion. Furthermore, this research highlights the importance of using data-driven approaches to reduce carbon emissions associated with transportation, enhance resource allocation at toll plazas, and promote sustainable highway transportation systems. Full article

Air Traffic Flow Management (ATFM) delay refers to the difference between a flight’s Target Take-Off Time (TTOT) and its Calculated Take-Off Time (CTOT), reflecting congestion levels in the air traffic network. ATFM delays are assigned to balance demand and capacity at key points in the network. The traditional First-Come, First-Served (FCFS) approach allocates delays strictly in the order flights are ready to depart, which is simple but inflexible. This study proposes a dynamic priority-based aircraft sequencing method at critical waypoints under multi-resource scenarios, aiming to reduce ATFM delays. An improved Constrained Position Shifting (CPS) constraint is introduced into the optimization model to enhance the influence of flight priority during decision-making. Additionally, three different priority strategies are designed to compare their respective impacts on ATFM delay. Finally, a dynamic priority-based ATFM delay optimization model is developed to address the identified challenges. Experimental results demonstrate that, compared with the FCFS scheme, the three priority strategies achieve maximum ATFM delay reductions of 30.5%, 44.1%, and 19.9%, respectively. The proposed model effectively allocates shorter delays to critical flights, optimizing resource utilization and improving the operational efficiency of the air route network. The research provides a reference framework for air traffic managers in allocating spatiotemporal resources across multiple congestion hotspots. By aligning priorities with network-wide efficiency goals, it overcomes traditional model limitations, avoids local optima, and supports globally optimal ATFM policy and practice. Full article

Against the background of traffic conflicts arising due to the highly concentrated population in high-density cities, this study aims to systematically identify the core transport needs of patients awaiting medical treatment; based on the theory of the Kano model, we construct a measurement system relating to patient transport needs when awaiting medical treatment that encompasses multiple levels. Taking 10 large general hospitals in Guangzhou as samples, this study collected data through questionnaires and auxiliary interviews, using the importance–sensitivity analysis method to accurately measure the degree of patient needs for each influencing factor of the transport environment for medical treatment. The study found that, among the primary needs (core basic needs), the perfection of public transport (which directly affects the convenience of medical care) is the core need with the highest degree of demand. Among the second-level needs (refined categorised demand levels), specifically relating to important attributes (essential needs), priority attention should be given to patient diversion, hospital–city connection, and corridor settings. As concerns the high value-added one-dimensional attributes (desired needs), focus should be placed on controlling health and safety distances and guiding the flow of medical treatment, while for high glamour attributes (glamour needs), primary consideration should be given to crowd distribution, stopping and resting, and direct access to the ground floor. The group difference analysis (grouped by emotional state, transport mode, and group type) showed that the first-level demand sensitivity ranking was highly consistent, and the second-level demand for urban connectivity, convenient transfer, and direct underground access were also common priorities. This study is the first to introduce the Kano model into the analysis of high-density urban healthcare transport systems, providing a clear basis for the grading of demand for the design of the transport environment for patients’ medical care. This is of great practical value for alleviating congestion and improving the resilience of emergency response in mega-cities in relation to medical care. Full article

With the development of dynamic wireless power transfer (DWPT) technology, the introduction of wireless charging lanes (WCLs) in traffic systems is seen as a promising trend for electrified transportation. Though there has been extensive discussion about the planning and allocation of WCLs in different situations, studies on traffic control models for WCLs are relatively lacking. Thus, this paper aims to design a coordinated optimization strategy for managing electric vehicle (EV) traffic on an expressway network, which integrates a corridor traffic flow model with a wireless power transmission model. Two components are considered in the control objective: the total energy increased for the EVs and the total number of EVs served by the expressway, over the problem horizon. By setting the trade-off coefficients for these two objectives, our model can be used to achieve mixed optimization of WCL traffic management. The decisions include metering of different on-ramps as well as routing plans for different groups of EVs defined by origin/destination pairs and initial SOC levels. The control problem is formulated as a novel linear programming model, rendering an efficient solution. Numerical examples are used to verify the effectiveness of the proposed traffic control model. The results show that with the properly designed traffic management strategy, a notable increase in charging performance can be achieved by compromising slightly the traffic performance while maintaining overall smooth operation throughout the expressway system. Full article

The rapid growth in the number of motor vehicles has exacerbated traffic congestion. The occurrence of congestion not only poses significant challenges for traffic management authorities but also severely impacts residents’ travel and daily routines. Against this backdrop, predicting traffic flow can provide crucial insights for anticipating changing traffic patterns. Therefore, this paper proposes a novel hybrid deep learning architecture (CNN-LSTM-GRU) for highway traffic flow prediction that integrates spatiotemporal and meteorological dimensions. Our approach constructs a multidimensional feature matrix encompassing temporal sequences, spatial correlations, and weather conditions. Convolutional Neural Networks (CNN) are employed to capture spatial patterns, while Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks jointly model temporal dependencies. Through systematic hyperparameter tuning and step-length optimization, we validate the model using real-world traffic data from a provincial highway network. The experimental evaluation analyzes the following two critical dimensions: (1) holiday vs. non-holiday traffic patterns, and (2) the impact of weather data integration. Comparative analysis reveals that our hybrid model demonstrates superior prediction accuracy over standalone LSTM, GRU, and their CNN-based counterparts (CNN-LSTM, CNN-GRU). Full article

Within urban tunnels, the lane boundary lines are typically solid, thereby prohibiting lane changes and overtaking. The establishment of a mobile operation zone in the slow lane can pose significant driving safety hazards not only to the slow lane within the tunnel but also to the middle and overtaking lanes at the tunnel exit. This article adopts the method of simulation of the establishment of an urban expressway three-lane VISSIM model, and selects the road traffic volume and speed of moving work zone as the independent variable parameters. Then, the influence range of a low-speed vehicle on the rear vehicles in the middle lane and slow lane and the traffic risk caused by a low-speed vehicle are analyzed. The results show that, irrespective of the variations in traffic volume and moving operation zone speed, the traffic flow within a 150 m range after the tunnel section was significantly influenced. This was because queuing and congested vehicles from the slow lane exited the tunnel, causing vehicles to change lanes and overtake in a concentrated manner. The moving operation zone has a substantial impact on the traffic flow in the slow lane. Under different moving operation zone speed conditions, the speed change trend of the following vehicles is consistent. When the moving operation zone speed was 5 km/h and the traffic volume exceeded 1200 pcu/h, the traffic flow behind the operation zone was directly affected, and within an observable longitudinal distance of 500 m, this impact did not dissipate. The research results can provide a scientific basis for the operation and management of urban tunnel low-speed vehicles. Full article

Generative artificial intelligence (GAI) has emerged as a transformative force in the Internet of Vehicles (IoV), addressing limitations of traditional AI such as reliance on large labeled datasets and narrow task applicability. This paper aims to systematically review recent advances in applying GAI to the IoV, with a focus on training, decision-making, and security. We begin by introducing the fundamental concepts of vehicular networks and GAI, laying the groundwork for readers to better understand the subsequent sections. Methodologically, we adopt a structured literature review, covering developments in synthetic data generation, dynamic scene reconstruction, traffic flow prediction, anomaly detection, communication management, and resource allocation. In particular, we integrate multimodal GAI capabilities with 5G/6G-enabled edge computing to support low-latency, reliable, and adaptive vehicular network services. Our synthesis identifies key technical challenges, including lightweight model deployment, privacy preservation, and security assurance, and outlines promising future research directions. This review provides a comprehensive reference for advancing intelligent IoV systems through GAI. Full article