Flow In Road Network Research Articles

Dynamic multi-scale spatial-temporal graph convolutional network for traffic flow prediction

This paper proposes a dynamic multi-scale spatial-temporal graph convolutional network (DS-STGCN) for traffic flow prediction. The network aims to comprehensively extract global and local dependencies in dynamic spatial-temporal data by inputting traffic network flow data to construct node feature graphs, topology graphs, and time slot feature graphs, capturing the complexity and dynamics of traffic flow. DS-STGCN interprets feature information of the traffic network from both spatial and temporal dimensions through dynamic multi-scale graph convolutional blocks. In the spatial dimension, these blocks use constraints at different levels to balance fine-grained local features and extensive global features, revealing the intrinsic structure of traffic flow data. In the temporal dimension, these blocks jointly learn with temporal convolutional blocks to capture multi-frequency time patterns and handle long sequence data, effectively extracting potential dependencies of time series. Furthermore, DS-STGCN effectively models the changing spatial-temporal relationships in road network flow by constructing dynamically adaptive updated adjacency tensors, generating dynamic graph structures to address the challenge of changing spatial-temporal relationships in the transportation system. Experimental results show that our method significantly outperforms other competing methods on five real traffic datasets (PEMS03, PEMS04, PEMS07, PEMS08 and METR-LA).

Capital-constrained maintenance scheduling for road networks considering traffic dynamics

Road network maintenance scheduling mainly considers budget limits in the previous studies and largely ignores the capital constraints. This study proposes a bi-objective mixed integer programming model, in which the net present value (NPV) is maximized and the increased total system travel time (ITSTT) due to maintenance activities under the capital constraints is minimized. Since the road network flows cannot reach an equilibrium state overnight due to the variation of network capacity, a link-based day-to-day dynamics model is developed to simulate the transient fluctuation in the traffic flows and calculate the total system travel time of each day. The bi-objective model is solved by the nondominated sorting genetic algorithm-II (NSGA-II) that generates a set of optimal Pareto solutions. The TOPSIS method is then adopted to determine the best compromise solution. Finally, a case study is conducted to demonstrate the effects of key parameters on the values of the two objectives.

Estimating Link Flows in Road Networks With Synthetic Trajectory Data Generation: Inverse Reinforcement Learning Approach

While traffic volume data from loop detectors have been the common data source for link flow estimation, the detectors only cover a subset of links. These days, other data sources such as vehicle trajectory data collected from vehicle tracking sensors are also incorporated. However, trajectory data are often sparse in that the observed trajectories only represent a small subset of the whole population, where the exact route sampling rate is unknown and may vary over space and time. In this paper, we develop a method that leverage these two limited data sources to enhance link flow estimation. This study proposes a novel generative modelling framework, where we formulate a vehicle’s link-to-link movements as a sequential decision-making problem using the Markov Decision Process framework. We propose an Inverse Reinforcement Learning-based method, based on which synthetic population vehicle trajectories can be generated to estimate link flows across the whole network. The proposed method ensures the generated population vehicle trajectories are consistent with the observed traffic volume and trajectory data. The proposed generative modelling framework is compared to two existing methods in a synthetic road network and validated in a real road network.

The impact of weather data on traffic flow prediction models

<span>Traffic flow prediction is an integral part of the intelligent transportation system (ITS) that helps in making well-informed decisions. Traffic flow prediction helps in alleviating traffic congestion as well as in some connected vehicles applications such as resources allocation. However, most of the existing models do not consider external factors such as weather data. Traffic flow in road networks is affected by weather conditions which affects the periodicity of traffic. These effects introduce some irregularity to the traffic pattern, making traffic flow prediction a challenging issue. In this paper, we present a detailed investigation on the impact of weather data on different traffic flow prediction models. The investigation presented in this paper demonstrates how adding weather data could improve the models’ prediction accuracy and efficiency.</span>

Guest Editorial Special Issue on Modeling Dynamic Transportation Networks in the Age of Connectivity, Autonomy and Data

The recent emergence of new technologies and systems such as connected and automated vehicles (CAVs), novel incentive and routing platforms, and shared mobility services is making a significant impact on traffic flow in road networks. The rapid development of these innovations, powered by new capabilities in data collection, communication, and vehicle autonomy raises both great opportunities and new challenges for managing and controlling the transportation network efficiently. It is thus imperative to integrate the emerging systems into a dynamic transportation network analysis, and to develop new methodologies, which coherently integrate dynamic traffic models with increasingly available data, and methods for large-scale computation. Consequently, they call for new theories, models, computational methods, and application scenarios to study dynamic transportation networks with the emerging technologies as essential components.

Dynamic Route Flow Estimation in Road Networks Using Data from Automatic Number of Plate Recognition Sensors

The traffic flow on road networks is dynamic in nature. Hence, a model for dynamic traffic flow estimation should be a very useful tool for administrations to make decisions aimed at better management of traffic. In fact, these decisions may in turn improve people’s quality of life and help to implement good sustainable policies to reduce the external transportation costs (congestion, accidents, travel time, etc.). Therefore, this paper deals with the problem of estimating dynamic traffic flows in road networks by proposing a model which is continuous in the time variable and that assumes the first-in-first-out (FIFO) hypothesis. In addition, the data used as model inputs come from Automatic Number of Plate Recognition (ANPR) sensors. This powerful data permits not only to directly reconstruct the route followed by each registered vehicle but also to evaluate its travel time, which in turn is also used for the flow estimation. In addition, the fundamental variable of the model is the route flow, which is a great advantage since the rest of the flows can be obtained using the conservation laws. A synthetic network is used to illustrate the proposed method, and then it is applied to the well-known Nguyen-Dupuis and Eastern Massachusetts networks to prove its usefulness and feasibility. The results on all the tested networks are very positive and the estimated flows reproduce the simulated real flows fairly well.

A Day-to-Day Stochastic Traffic Flow Assignment Model Based on Mixed Regulation

In many cases, the final path selection of travellers’ is not the shortest path, due to the limited computing power and high cost of path search. To solve the problem, this paper proposes a day-to-day (DTD) stochastic traffic flow assignment model that regulates the traffic flow based on the travel time (travel cost) and residual congestion of optional paths. The regulation mechanism is called the mixed regulation. Then, the authored proved the existence, uniqueness and stability of the model solution. The proposed model was verified through simulation on a Nguyen-Dupuis road network. The results show that traffic flows and travel times of all paths reached the equilibrium state, thanks to the DTD mixed regulation for 20~30 days. From the traffic flows and congestion degrees of different sections, it can be seen that our model with mixed regulation diverts the traffic flow to the sections with a low congestion degree, and encourages travellers to drive through the sections with a low traffic flow. In addition, the congestion degrees of the four most congested sections decreased by 5.8%, 4%, 7% and 1.2%, respectively, and the entire road network exhibited a slight downward trend in mean congestion degree. These results prove that our model can uniformize the traffic flow, improve the operation efficiency and alleviate the congestion of the road network. These findings shed new light on the control, guidance and planning of traffic flow in road networks.

Application of network-level traffic flow model to road maintenance in China

This paper shows how to simulate highway traffic flows using a network-level traffic flow model (NTFM) for both urban and motorway road networks. It highlights how road maintenance works can be modelled using a roadwork node in the network and how the modelling can be used to predict the effects of maintenance works on the level of service of the highway network. In particular, the paper aims to show how the theory of vehicle flow fluctuation can be used within the roadwork node for better representation of reality while modelling the road network flow. The NTFM is illustrated using traffic characteristics and road maintenance works on a motorway and trunk highway network in China. It is shown how the proposed method is useful in comparing worksite arrangements and describing the effects on flow rates due to roadworks.

Optimal control of urban air pollution related to traffic flow in road networks

Air pollution is one of the most important environmental problems nowadays. In large metropolitan areas, the main source of pollution is vehicular traffic. Consequently, the search for traffic measures that help to improve pollution levels has become a hot topic today. In this article, combining a 1D model to simulate the traffic flow over a road network with a 2D model for pollutant dispersion, we present a tool to search for traffic operations that are optimal in terms of pollution. The utility of this tool is illustrated by formulating the problem of the expansion of a road network as a problem of optimal control of partial differential equations. We propose a complete algorithm to solve the problem, and present some numerical results obtained in a realistic situation posed in the Guadalajara Metropolitan Area (GMA), Mexico.

Topology control of VANET based on the multi-radio in vehicle and traffic flow in road network

Topology control of VANET based on the multi-radio in vehicle and traffic flow in road network

A statistical method for estimating predictable differences between daily traffic flow profiles

It is well known that traffic flows in road networks may vary not only within the day but also between days. Existing models including day-to-day variability usually represent all variability as unpredictable fluctuations. In reality, however, some of the differences in flows on a road may be predictable for transport planners with access to historical data. For example, flow profiles may be systematically different on Mondays compared to Fridays due to predictable differences in underlying activity patterns. By identifying days of the week or times of year where flows are predictably different, models can be developed or model inputs can be amended (in the case of day-to-day dynamical models) to test the robustness of proposed policies or to inform the development of policies which vary according to these predictably different day types. Such policies could include time-of-day varying congestion charges that themselves vary by day of the week or season, or targeting public transport provision so that timetables are more responsive to the day of the week and seasonal needs of travellers. A statistical approach is presented for identifying systematic variations in daily traffic flow profiles based on known explanatory factors such as the day of the week and the season. In order to examine day-to-day variability whilst also considering within-day dynamics, the distribution of flows throughout a day are analysed using Functional Linear Models. F-type tests for functional data are then used to compare alternative model specifications for the predictable variability. The output of the method is an average flow profile for each predictably different day type, which could include day of the week or time of year. An application to real-life traffic flow data for a two-year period is provided. The shape of the daily profile was found to be significantly different for each day of the week, including differences in the timing and width of peak flows and also the relationship between peak and inter-peak flows. Seasonal differences in flow profiles were also identified for each day of the week.

A multi-level and modular model for simulating the urban flooding and its application to Tianjin City

In this study, we present a multi-level and modular model for urban flood. It consists of four modules which simulate: (1) two-dimensional (2D) flow routing on communities, (2) 2D flow routing on road networks, (3) 2D flow routing on river network, and (4) one-dimensional (1D) flow routing on sewer system. The four modules are coupled by pumping stations, gutter inlets, and procedures. The gates or pumping stations connect the sewer system and the river network, allowing volume exchange between underground pipes and river network. The gutter inlets connect the sewer system and the river network, allowing volume exchange between the road network and underground pipes. The consistent treatment has also been included between the community areas and sewer system. The coupling among the four modules not only captures the interaction between surface and subsurface components, but also divides the surface open channels into community areas flow, road network flow and river network flow. Therefore, it greatly improves the calculation efficiency and accuracy. A catchment in Hexi District of Tianjin City, China, is used as the study case to validate the model. The simulated results agree well with the measurements in the following four aspects: the water levels of the Jin River, water depth at feature points, waterlog conditions in two important areas, and water balance. This model, which greatly improves the simulation of the dual drainage system, can be very useful for future prediction and identification of urban flooding.

Structure of complex networks: Quantifying edge-to-edge relations by failure-induced flow redistribution

AbstractThe analysis of complex networks has so far revolved mainly around the role of nodes and communities of nodes. However, the dynamics of interconnected systems is often focalized on edge processes, and a dual edge-centric perspective can often prove more natural. Here we present graph-theoretical measures to quantify edge-to-edge relations inspired by the notion of flow redistribution induced by edge failures. Our measures, which are related to the pseudo-inverse of the Laplacian of the network, are global and reveal the dynamical interplay between the edges of a network, including potentiallynon-localinteractions. Our framework also allows us to define the embeddedness of an edge, a measure of how strongly an edge features in the weighted cuts of the network. We showcase the general applicability of our edge-centric framework through analyses of the Iberian power grid, traffic flow in road networks, and theC. elegansneuronal network.

Signal Setting Optimisation with Route Choice Behaviour and Acyclic Profiles of Vehicular Flows in Road Networks

Traffic signal setting is an important element in traffic control and in the management of urban road networks both in ordinary conditions (ex. recurrent traffic congestion) and/or in emergency conditions (ex. non-recurrent traffic congestion caused by unexpected events). The paper presents models and procedures for signal setting optimisation of signalised junctions in a congested road network. Two interacting procedures are developed to solve the system of models:•an optimisation procedure, based on a genetic algorithm, to obtain an optimal configuration of signal setting parameters and•a simulation procedure, incorporating a path choice model with explicit path enumeration and a flow propagation model, to capture effects of signal setting configuration on user path choice behaviour.The main research contributions of this work are related to: (i) the modelling of path choice behaviour in a dynamic context with vehicular flow propagation of on link and spill-back; (ii) the optimisation procedure of signal setting parameters. The system of models and procedures has been applied and validated on a real experimental test site, where the transport system is in dynamic conditions due to time-varying road network and travel demand.

Internal catchment process simulation in a snow‐dominated basin: performance evaluation with spatiotemporally variable runoff generation and groundwater dynamics

AbstractHydrologic models have increasingly been used in forest hydrology to overcome the limitations of paired watershed experiments, where vegetative recovery and natural variability obscure the inferences and conclusions that can be drawn from such studies. Models are also plagued by uncertainty, however, and parameter equifinality is a common concern. Physically‐based, spatially‐distributed hydrologic models must therefore be tested with high‐quality experimental data describing a multitude of concurrent internal catchment processes under a range of hydrologic regimes. This study takes a novel approach by not only examining the ability of a pre‐calibrated model to realistically simulate watershed outlet flows over a four year period, but a multitude of spatially‐extensive, internal catchment process observations not previously evaluated, including: continuous groundwater dynamics, instantaneous stream and road network flows, and accumulation and melt period spatial snow distributions. Many hydrologic model evaluations are only on the comparison of predicted and observed discharge at a catchment outlet and remain in the ‘infant stage’ in terms of model testing. This study, on the other hand, tests the internal spatial predictions of a distributed model with a range of field observations over a wide range of hydroclimatic conditions. Nash‐Sutcliffe model efficiency was improved over prior evaluations due to continuing efforts in improving the quality of meteorological data collection. Road and stream network flows were generally well simulated for a range of hydrologic conditions, and snowpack spatial distributions were well simulated for one of two years examined. The spatial variability of groundwater dynamics was effectively simulated, except at locations where strong stream–groundwater interactions exist. Model simulations overall were quite successful in realistically simulating the spatiotemporal variability of internal catchment processes in the watershed, but the premature onset of simulated snowmelt for one of the simulation years has prompted further work in model development. Copyright © 2011 John Wiley & Sons, Ltd.

Urban Road Network Traffic Congestion Prediction Model Based on Probe Vehicle Technology

AbstractA model for urban road network traffic congestion forecast based on probe vehicle technology, fuzzy logic judgement and back-propagation (BP) neural network was proposed. A three-layer BP neural network model was built to estimate the real time traffic flow of road network and to obtain BP neural network training specimen for the training by probe vehicle data and video data. Then the congestion possibility, level of congestion and the forming time of the link were estimated based on the road network topology and multifile fuzzy logic reasoning. The in-situ test shows good forecast result by the model.

A hybrid macroscopic-based model for traffic flow in road networks

Recently, the unquestionable growth of interest to increase the operational efficiency and capability of transportation systems led to the development of a large number of traffic modeling theories. One of the major operational issues when developing a transportation system management model lies in the selection of the appropriate methodological approach with respect to several decisions, such as the selection of the type of input and output data as well as the qualitative representation and the computational power of the model. Despite the considerable effort in the area, there is still not an approach which per se models effectively the various dynamically evolving features of traffic in road networks. The present paper addresses this issue by introducing a new hybrid approach which combines the complementary features and capabilities of both continuum mathematical models e.g. [1,6,23,26] and knowledge-based models e.g. [7,22,28] in order to describe effectively traffic flow in road networks.

Forecasting model of short-term traffic flow for road network based on independent component analysis and support vector machine

Traffic flow forecasting is one of the important issues for the research of Intelligent Transportation System(ITS).Through analyzing the characteristics of data collected by different observation place on the same road,the authors proposed a new prediction method of short-term traffic flow in road network based on Independent Component Analysis(ICA)and Support Vector Machine(SVM).First,the traffic flow data of every observation point on the same road was turned into independent source signal through ICA method.Second,SVM model was used to train and predict the source signal,and through Genetic Algorithm(GA)parameters were optimized.At last,the traffic flow forecasting data were obtained by an inverse transform.Real traffic data were applied to test the proposed prediction model.The experimental results show that this method not only is more accurate than the method which uses SVM directly to predict traffic flow,but also can get rid of the data interaction of every observation points on the same road.

Improved Algorithm about the Maximum Flow in Road Network Using Auxiliary Graph Theory

In road network planning, locating the key sections will be very useful to solve the traffic bottleneck. Locating the key sections is equal to finding the maximum flow. To begin with, this paper analyzed and compared kinds of methods solving the network maximum flow problem. Second, this paper pointed out a convenient method based on the auxiliary graph theory and Dijkstra method. That is to say, the method could be used to get the minimum cutset and the maximum flow using the shortest path algorithm, and the VC++ program was also used. Finally, based on the road network of Tianjin, an example was given for further explanation. Through comparison, we know that it is a convenient method to locate the key sections and can support the transportation planning and management with theoretical basis and data.