Traffic Flow Distribution Research Articles

An Adaptive Framework for Traffic Congestion Prediction using Deep Learning

Aim and background: Congestion on China's roads has worsened in recent years due to the country's rapid economic development, rising urban population, rising private car ownership, inequitable traffic flow distribution, and growing local congestion. As cities expand, traffic congestion has become an unavoidable nuisance that endangers the safety and progress of its residents. Improving the utilization rate of municipal transportation facilities and relieving traffic congestion depend on a thorough and accurate identification of the current state of road traffic and necessitate anticipating road congestion in the city. Methodology: In this research, we suggest using a deep spatial and temporal graph convolutional network (DSGCN) to forecast the current state of traffic congestion. To begin, we grid out the transportation system to create individual regions for analysis. In this work, we abstract the grid region centers as nodes, and we use an adjacency matrix to signify the dynamic correlations between the nodes. Results and Discussion: The spatial correlation between regions is then captured utilizing a Graph Convolutional-Neural-Network (GCNN), while the temporal correlation is captured using a two-layer long and short-term feature model (DSTM). Conclusion: Finally, testing on real PeMS datasets shows that the DSGCN has superior performance than other baseline models and provides more accurate traffic congestion prediction.

Collaborative Optimization Framework for Coupled Power and Transportation Energy Systems Incorporating Integrated Demand Responses and Electric Vehicle Battery State-of-Charge

The growing adoption of electric vehicles (EVs) and advancements in dynamic wireless charging (DWC) technology have strengthened the interdependence between power distribution networks (PDNs) and electrified transportation networks (ETNs), leading to the emergence of coupled power and transportation energy systems (CPTESs). This development introduces new challenges, particularly as DWC technology shifts EV charging demand from residential plug-in charging to charging-while-driving during commuting hours, causing simultaneous congestion in both ETNs and PDNs during peak times. The present work addresses this issue by developing a collaborative optimization framework for CPTESs that incorporates integrated demand responses (IDRs) and EVs battery state-of-charge (SOC). In the ETN, a multiperiod traffic assignment model with time-shiftable traffic demands (MTA-TSTD) is established to optimize travelers’ routes and departure times while capturing traffic flow distribution. Meanwhile, effective path generation models with EVs battery SOC are proposed to optimize charging energy during driving and construct the effective path sets for MTA-TSTD. In the PDN, a multiperiod optimal power flow model with time-shiftable power demands (MOPF-TSPD) is formulated to schedule local generators and flexible power demands while calculating the power flow distribution. To enhance temporal and spatial coordination in CPTESs, a distributed coordinated operation model considering IDRs is proposed, aiming to optimize energy consumption, alleviate congestion, and ensure system safety. Finally, an adaptive effective path generation algorithm and an ETN–PDN interaction algorithm are devised to efficiently solve these models. Numerical results on two test systems validate the effectiveness of the proposed models and algorithms.

Measurement-Noise Filtering for Automatic Discovery of Flow Splitting Ratios in ISP Networks

Network telemetry and analytics is essential for providing highly dependable services in modern computer networks. In particular, network flow analytics for ISP networks allows operators to inspect and reason about traffic patterns in their networks in order to react to anomalies. High performance network analytics systems are designed with scalability in mind, and can consequently only observe partial information about the network traffic. Still, they need to provide a holistic view of the traffic, including the distribution of different traffic flows on each link. It is impractical to monitor such fine-grained telemetry, and in large, heterogeneous networks it is often too complex and error-prone, if not impossible, to access and maintain all technical specifications and router-specific configurations needed to determine e.g. the load balancing weights used when traffic is split onto multiple paths. The ratios by which flows are split on the possible paths must be derived indirectly from the measured flow demands and link utilizations. Motivated by a case study provided by a major European ISP, we suggest an efficient method to estimate the flow splitting ratios. Our approach, based on quadratic linear programming, is scalable and achieves robustness to the measurement noise found in a typical network analytics deployment by filtering out certain constraints in the linear program. Finally, we implement an automated tool for estimating the flow splitting ratios and document its applicability on real data from the ISP.

Predicting the urban stormwater drainage system state using the Graph-WaveNet

Graph Neural Networks (GNNs) have been applied to network data such as traffic flow and water distribution systems, yet their use in predicting the state of urban stormwater drainage systems remains rare. This study investigates the application of Graph-WaveNet (GWN), a type of GNN, in forecasting the states of stormwater systems in Kowloon, Hong Kong. Data was sourced from the Storm Water Management Model (SWMM) spanning 43 rainfall events from 2020 to 2023. Based on the preceding 30 to 60 min of network states and rainfall data, GWN predicted junction inflows, pipe flow rates, and relative water depths (fraction of full area filled by flow) for lead times up to 20, 20, and 30 min, with an R2 greater than 0.6, respectively. Prediction accuracy declines with longer forecast horizons. GWN predicts more time steps ahead for pipes’ flow rates and junctions’ inflows, but fewer for relative water depths during peak versus non-peak periods. It is also more effective at predicting states of large pipes and connected junctions downstream, compared to smaller upstream components. GWN's accuracy improves significantly with precise rainfall nowcasting inputs. This study establishes a significant baseline for GWN's performance in predicting urban stormwater systems during rainfall events.

Flow prediction of mountain cities arterial road network for real-time regulation

This study aims to enhance the understanding of vehicle path selection behavior within arterial road networks by investigating the influencing factors and analyzing spatial and temporal traffic flow distributions. Using radio frequency identification (RFID) travel data, key factors such as travel duration, route familiarity, route length, expressway ratio, arterial road ratio, and ramp ratio were identified. We then proposed an origin–destination path acquisition method and developed a route-selection prediction model based on a multinomial logit model with sample weights. Additionally, the study linked the traffic control scheme with travel time using the Bureau of Public Roads function—a model that illustrates the relationship between network-wide travel time and traffic demand—and developed an arterial road network traffic forecasting model. Verification showed that the prediction accuracy of the improved multinomial logit model increased from 92.55 % to 97.87 %. Furthermore, reducing the green time ratio for multilane merging from 0.75 to 0.5 significantly decreased the likelihood of vehicles choosing this route and reduced the number of vehicles passing through the ramp. The flow prediction model achieved a 97.9 % accuracy, accurately reflecting actual volume changes and ensuring smooth operation of the main airport road. This provides a strong foundation for developing effective traffic control plans.

Efficiency and fuel consumption of mixed traffic flow with lane management of CAVs

The mixed traffic flow of connected and automated vehicles (CAVs) and human-driven vehicles (HDVs) will exist on highways for a long time, as the deployment of CAVs is gradual. To reduce the negative impact of HDVs on CAVs, the deployment of dedicated lanes has been considered an effective solution. Along with the dedicated lanes, three different lane management strategies will be formed, which are (C, H) strategy (CAVs dedicated lanes and HDVs dedicated lanes), (C, G) strategy (CAVs dedicated lanes and general lanes), and (G, H) strategy (general lanes and HDVs dedicated lanes). To evaluate the influence of dedicated lane settings on mixed traffic flow comprehensively, this paper proposes a framework for evaluating road segment efficiency and fuel consumption by considering lane management strategies. First, the possible traffic flow equilibrium states under three lane management strategies are discussed, and the characteristics of five car-following modes in mixed traffic flow are analyzed. Then, a mixed traffic flow capacity model considering platoon size is introduced to the traditional BPR function to establish a speed estimation model for mixed traffic flow and a fuel consumption estimation model for mixed traffic flow. Next, the traffic flow distribution model at the lane level in a steady state is derived for different lane management strategies. Based on the traffic flow distribution model, the speed estimation model and the fuel consumption estimation model for mixed traffic flow, which consider lane management strategies, are proposed. Finally, a numerical simulation is conducted to analyze the effects of different lane management strategies and configuration schemes on road segment efficiency and fuel consumption. The results of numerical experiments show that (1) at the same traffic demand, the operational speeds of vehicles under the (C, H) strategy and (G, H) strategy tend to increase and then decrease with the increase in the penetration rate of CAVs. While the speed of the vehicle under the (C, G) strategy increases with the increase in the penetration rate of CAVs. (2) Compared with the baseline strategy, all three management strategies can improve the operating efficiency of vehicles under certain traffic conditions. (3) At the same traffic demand, the average fuel consumption under the three strategies tends to decrease first and then increase slightly as the penetration rate increases. Increasing the number of dedicated lanes under specific traffic conditions can significantly increase the fuel consumption reduction rate under each strategy. At the same penetration rate, this advantage diminishes with the increase in traffic demand. (4) The increase in platoon size favors the efficiency of vehicle operations under different strategies. However, as platoon size increases, the marginal benefit of increasing platoon size becomes smaller and smaller. In addition, the average fuel consumption of vehicles has a low sensitivity to platoon size, and increasing platoon size may not always reduce fuel consumption.

A traffic-weather generative adversarial network for traffic flow prediction for road networks under bad weather

Traffic flow prediction is pivotal in providing reliable information for intelligent traffic systems. Unexpected events, such as bad weather, unavoidably impact the precision of traffic flow prediction. Therefore, to achieve accurate traffic flow prediction results in road networks under bad weather, a novel Traffic-Weather Generative Adversarial Network (TWeather-GAN model) is developed. This model comprises a Generator and a Discriminator. The Generator incorporates both the traffic and weather modules to extract the spatiotemporal patterns hidden in traffic flow and weather data. In the traffic and weather modules, the gated convolutional layer, Encoder-Decoder architecture, and attention mechanism are established. In the Discriminator, the gated convolutional layer and bidirectional long short-term memory neural network are introduced. Traffic flow data under fog, strong wind, and heavy rain are selected to test the seven baseline models and the proposed model, and the ablation experiments are conducted to analyze the mechanism of the proposed model. The experiments demonstrate that the TWeather-GAN model outperforms the baseline models under bad weather, and makes the prediction error have an average reduction of 0.20%–34.81%, 3.21%–35.22%, and 9.46%–39.10%, respectively, under one-step prediction, three-step prediction, and six-step prediction. Furthermore, establishing the gated convolutional layer and the weather module enhances the accuracy of traffic flow prediction under bad weather. Results show that traffic flow fluctuations and distributions differ under fog and strong wind, and heavy rain affects the trend of traffic flow over one day.

A Vehicle-to-Grid planning framework incorporating electric vehicle user equilibrium and distribution network flexibility enhancement

The rapid surge in electric vehicle (EV) adoption, coupled with advancements in charging technologies, emphasizes the critical necessity for expanding EV recharging infrastructure. Simultaneously, the Distribution Network (DN) encounters escalating challenges in meeting charging demand during peak traffic periods. Consequently, there is a mounting demand for the deployment of innovative Vehicle-to-Grid (V2G) technologies to augment the DN’s flexibility in power dispatch and alleviate travel costs for EV users. Hence, this paper proposes an EV-user-equilibrium-(UE)-constrained V2G planning framework that enhances flexibility in the DN. The framework aims to ascertain the optimal placement and capacity of EV charging stations (EVCSs) and V2G charging piles within the Transportation Network (TN). It takes into account the equilibrium condition stemming from competitive EV charging and routing behaviors alongside the optimal expansion of DN energy resources to accommodate the electricity supplied by the V2G piles. This study commences by analyzing EV drivers’ travel decisions, considering the influence of charging and V2G pile locations and sizes. Subsequently, we tackle the Traffic Assignment Problem with User Equilibrium (TAP-UE) model to characterize the steady-state traffic flow distribution of EVs. Following this, we formulate the optimization model for the Coordinated Power and Transportation Network (CPTN), which encompasses the optimal expansion of DN facilities and traffic flow regulation under UE conditions. To mitigate the computational complexity associated with the V2G planning model, we introduce a series of linearization methods to obtain a manageable Mixed-Integer Linear Programming (MILP) solution. Finally, to validate the efficacy of our proposed planning framework, we apply it to two test systems, including a real-world case study. Through these case studies, we explore the necessity and potential benefits of V2G technologies.

Optimizing gate control coordination signal for urban traffic network boundaries using multi-agent deep reinforcement learning

The macro-aggregated dynamic characteristics of urban network traffic flow encapsulated embodied in macroscopic fundamental diagram theory provide a concise perspective for network-level traffic control. However, accurately translating inter-regional macro-traffic flow control into fine-grained control at each intersection along the boundary remains challenging. To address this issue, a regional boundary intersections collaborative control method is proposed, leveraging the model-free data-driven advantages of multi-agent deep reinforcement learning. First, the proposed model develops a real-time communication module using a bi-directional long short-term memory model and an attention mechanism, enabling intersections to capture spatiotemporal features of the boundary traffic flow. Second, the feedback reward mechanism of the boundary intersection agent is meticulously constructed by considering the traffic state of the protected region, the boundary control pressure, and the traffic pressure of the intersection. The proposed model fosters cooperation at boundary intersections and delicately balances their macro-micro control contradictions based on mastering the boundary state information of the road network. Data-driven boundary gating control can fine-regulate the traffic flow distribution of the network from the perspective of macro-micro to promote its operation efficiency. The simulation experiments verify the effectiveness of the proposed multi-agent signal collaborative control method.

Fatigue life prediction of corroded hangers with parallel steel wire in a network arch bridge under vehicle-bridge interaction based on a novel noniterative simplified method

Fatigue damage is a main concern during the life-cycle maintenance of bridges with corroded hangers under vehicle-bridge interaction (VBI). In this study, a novel noniterative simplified method for VBI is proposed to analyze the responses of hangers in a network arch bridge, considering the effects of vehicle types, driving lanes, and road surface roughness. Furthermore, the fatigue life of corroded hangers is investigated, considering the actual traffic flow and uneven distribution of stress in the hanger’s cross-section. The results show that longitudinal and transverse bending moments of the hanger significantly affect the stress of steel wires in one hanger but have limited influence on stress amplitudes. The fatigue lives of steel wires are considerably reduced when subjected to both uniform and pitting corrosion. The fatigue life of steel wires is reduced under deteriorating road surface conditions. Moreover, as the fatigue life of steel wires in the hangers declines, the variations in fatigue life among individual steel wires within a single hanger gradually decrease. Consequently, this trend leads to a heightened likelihood of continuous wire breakage.

Study of the correspondences distribution of vehicle traffic on the road network of cities

The problems of traffic management in large cities are complicated as the increase in traffic volume far exceeds the road network capacity. It leads to saturation of the road network, which negatively affects its functioning. This article analyzes the state of the art of improving the quality of road traffic by predicting the size of traffic after the implementation of traffic management measures. While the issue of modeling traffic flow parameters based on technological factors has been sufficiently studied, the problems of considering the human factor need to be clarified. The object of study is traffic flow on the city's road network. The criteria used by drivers to compare the characteristics of alternative traffic routes were identified based on the field study results. Based on the data obtained, the parameters of alternative routes that drivers choose when driving on the road network are formed. According to the survey results, the most significant factor is the minimum mileage along the route. Deviations from the shortest route were determined to determine the patterns of distribution of correspondence of non-route vehicles along alternative routes. It made it possible to define the distribution of transport correspondence along alternative routes. Since the process under consideration is probabilistic, the law of distribution of a random variable was determined. After having determined the law of distribution of a random variable for the data obtained, the calculations showed that the change in the random variable is well described by the gamma distribution. It was determined that the share of transport correspondence that will be carried on them decreases as the length of the route deviates from the shortest one. The obtained results make it possible to predict road network congestion by modeling the distribution of traffic flows. In further research, it would be advisable to analyze the distribution of transport correspondence by other criteria.

Data-driven traffic sensor location and path flow estimation using Wasserstein metric

This paper introduces link information value obtained by the traffic sensors and presents a traffic sensor location and flow estimation joint optimization model in an urban road network. In contrast to most previous studies, this paper adds new traffic sensors into the existing sensor network and proposes a data-driven path flow measurement method based on Wasserstein metric, which is utilized to measure the distance between the estimated traffic flow distribution and the actual distribution. Furthermore, this paper develops a customized greedy algorithm by combining a search strategy for the link information value to obtain the optimal sensor location scheme and perform traffic flow estimation under different budget conditions. Numerical experiments are conducted on Sioux-Falls test network and Eastern Massachusetts interstate highway subnetwork to verify the accuracy and effectiveness of the proposed model based on Wasserstein metric and the developed solution method. Computational results show that the sensor location scheme generated by the model based on Wasserstein metric can reduce the estimation error of the traffic flow compared with KL divergence model under the same deployment cost. Additionally, the customized greedy algorithm can achieve the better performance than the Brute force algorithm in terms of computing time and solution quality.

Effect of road pricing on the spatial distribution of traffic flow in a grid network

This paper develops a continuous approximation model for analyzing the effect of road pricing on the spatial distribution of traffic flow. The traffic flow density, which describes traffic flow as a function of position, is derived for a rectangular city with a grid network. The analytical expression for the traffic flow density demonstrates how the toll level, the size and shape of the toll area, and the shape of the city affect the spatial distribution of traffic flow. As the size of the toll area increases, reducing the traffic flow density at the city center becomes difficult. As the aspect ratio of the toll area increases, the traffic flow density at the city center increases. The shape of the city has less impact on the traffic flow density than the shape of the toll area.

Road Network-Guided Fine-Grained Urban Traffic Flow Inference.

Accurate inference of fine-grained traffic flow from coarse-grained one is an emerging yet crucial problem, which can help greatly reduce the number of the required traffic monitoring sensors for cost savings. In this work, we note that traffic flow has a high correlation with road network, which was either completely ignored or simply treated as an external factor in previous works. To facilitate this problem, we propose a novel road-aware traffic flow magnifier (RATFM) that explicitly exploits the prior knowledge of road networks to fully learn the road-aware spatial distribution of fine-grained traffic flow. Specifically, a multidirectional 1-D convolutional layer is first introduced to extract the semantic feature of the road network. Subsequently, we incorporate the road network feature and coarse-grained flow feature to regularize the short-range spatial distribution modeling of road-relative traffic flow. Furthermore, we take the road network feature as a query to capture the long-range spatial distribution of traffic flow with a transformer architecture. Benefiting from the road-aware inference mechanism, our method can generate high-quality fine-grained traffic flow maps. Extensive experiments on three real-world datasets show that the proposed RATFM outperforms state-of-the-art models under various scenarios. Our code and datasets are released at https://github.com/luimoli/RATFM.

РЕШЕНИЕ ПРОБЛЕМ ТРАНСПОРТНЫХ ЗАТОРОВ СРЕДСТВАМИ СВЕТОФОРНОГО РЕГУЛИРОВАНИЯ

The analysis of methods for solving the problem of traffic congestion, including approaches to operational management of the distribution of traffic flows by means of informing users and traffic light regulation in order to prevent traffic congestion and minimize their consequences, is presented. The method of balancing transport demand and network capacity with traffic light reg-ulation and the results of its application on the example of Penza are considered. The positive effect of its application and the need for an integrated approach to solving the problem of traffic congestion are shown.

The impact of highway construction on the volume and distribution of traffic flows: Case study for highway A2 in Serbia

In this paper, the total traffic flows on the old and new road were analyzed, after the construction of the new road. Traffic flows were analyzed on the example of the A2 highway (the Miloš Veliki highway) and the B-22 road (the so-called Ibarska Magistrala). The aim of the papaer is the analysis of the distribution of total traffic flows, as well as the analysis of traffic flows by toll categories between old and new road. Statistical analysis methods were applied in the paper. The obtained results show a significant increase in total traffic flows after the construction of the highway on the Belgrade - Čačak road direction. The analysis yielded very interesting data on the distribution of traffic flows. In further research, it is necessary to conduct a survey in order to reveal the realtions and influencing factors on the volume and distribution of traffic flows, as well as the category of traffic flows (newly created, diverted and normal traffic flows).

Computation Offloading for Energy and Delay Trade-Offs With Traffic Flow Prediction in Edge Computing-Enabled IoV

An unprecedented prosperity in artificial intelligence promotes the development of Internet of Vehicles (IoV). Assisted by edge computing, vehicles enable to offload data to edge servers in close proximity to users for processing, thus making up for the shortage of local computing resources. However, due to the uneven space-time distribution of traffic flow, edge servers of a certain road segment may be overwhelmed by the surge of service requests. Furthermore, IoV system will incur significant additional energy consumption and time delay because of the absence of a proper computation offloading scheme between edge servers. To cope with above challenges, a computing offloading method for energy and delay trade-offs with traffic flow prediction in edge computing-enabled IoV is proposed. We first design the graph weighted convolution network (GWCN) that can fully excavate the connectivity and distance relation information between road segments to conduct traffic flow prediction. The short-term prediction results are utilized as the basis for adjusting the resource allocation of edge resources in different regions. Then, a computation offloading method driven by deep deterministic policy gradient (DDPG) is leveraged to obtain an optimal computation offloading scheme for edge servers. Finally, extensive comparative experiments demonstrate the low prediction error of GWCN and superior performance of DDPG-driven method in reducing total time delay and energy consumption.

On Accelerated Coordinate Descent Methods for Searching Equilibria in Two-Stage Transportation Equilibrium Traffic Flow Distribution Model

On Accelerated Coordinate Descent Methods for Searching Equilibria in Two-Stage Transportation Equilibrium Traffic Flow Distribution Model

Road criticality assessment to improve commutes during floods

Flooding undermines the functionality of road links, leading to commute disruption, such as increased commuting time and immobility. Therefore, identifying the critical roads under different flood scenarios is of great significance for transportation planning to minimize flood risk. Previous studies on the road criticality under pluvial floods have focused on mostly structural characteristics of the roads, lacking the quantification of the flood impact on traffic. We employed an integrated framework coupling pluvial flood modeling with commute simulation to assess the criticality of flooded road links under three scenarios (10, 30, and 50 year floods). The output of the flood modeling was used to reconstruct the road network. The routes of each individual commuter under a business as usual (BAU) scenario and three flood scenarios were obtained from the commute simulation. As the flood return period increased, the distribution of traffic flow became more uneven. The road criticality assessment revealed that greater flood impact on roads was closely related to the proximity of water bodies, low-lying topography, and greater commute demands of the region and provided evidence to prioritize locations for interventions. This study helps understand the resilience of transportation systems during floods, thereby enabling practitioners to design effective strategic plans in terms of emergency response, risk management, and urban planning.

Measuring and modeling the effects of green barriers on the spatial distribution of fine particulate matter at roadside

Green barriers are regarded as a viable strategy for improving road air quality, and yet there lacks consensus regarding how they can actively regulate air pollution, particularly under dynamic traffic, meteorology and built environments. To address this, field measurements were conducted to assess the levels of fine particulate matter (PM2.5) before and after green barriers along an urban expressway. The measurements revealed distinct spatial attenuation patterns in vegetation-rich and vegetation-sparse areas. Specifically, the first-layer green barriers were found to significantly reduce PM2.5 levels by 8.4–9.3% in the vegetation-rich area. The magnitude of PM2.5 reduction at all locations distant from the motorway varied depending on the wind direction. On average, there are reductions of 7.8% (winter) and 13.0% (summer) in PM2.5 levels under cross-road winds. A validated hybrid model further demonstrated the significant impact of green barriers on pollutant diffusion, showing an unstable spatial variation trend under changing wind conditions. A combination of trees and hedgerows with a tree spacing of 3 m and obovate or spherical tree crowns proved effective in reducing roadside pollution. Simulations also suggested significant pollution reduction when considering appropriate spatial distribution of traffic flow and solid barrier alongside green barriers. These findings underscore the potential of combining green barriers with diverse measures to maximize the reduction of traffic-related pollution at roadside.