Search

With the promotion of ChatGPT to the public, Large language models indeed showcase remarkable common sense, reasoning, and planning skills, frequently providing insightful guidance. These capabilities hold significant promise for their application in urban traffic management and control. However, large language models (LLMs) struggle with addressing traffic issues, especially processing numerical data and interacting with simulations, limiting their potential in solving traffic-related challenges. In parallel, specialized traffic foundation models exist but are typically designed for specific tasks with limited input-output interactions. Combining these models with LLMs presents an opportunity to enhance their capacity for tackling complex traffic-related problems and providing insightful suggestions. To bridge this gap, we present TrafficGPT—a fusion of multiple LLMs and traffic foundation models. This integration yields the following key enhancements: 1) empowering LLMs with the capacity to view, analyze, process traffic data, and provide insightful decision support for urban transportation system management; 2) facilitating the intelligent deconstruction of broad and complex tasks and sequential utilization of traffic foundation models for their gradual completion; 3) aiding human decision-making in traffic control through natural language dialogues; and 4) enabling interactive feedback and solicitation of revised outcomes. By seamlessly intertwining large language model and traffic expertise, TrafficGPT not only advances traffic management but also offers a novel approach to leveraging AI capabilities in this domain. The TrafficGPT demo can be found in https://github.com/lijlansg/TrafficGPT.git. • We propose the TrafficGPT framework, which combines large language models with expertise in traffic management. • TrafficGPT empowers LLMs to observe, analyze, process traffic data, and offer insightful decisions for traffic management. • We compare the proposed model with ChatGPT-4 Data Analyst to highlight its response accuracy and efficiency. • The adaptability and stability of TrafficGPT across various tasks, utilizing different foundational LLMs, are validated. [ABSTRACT FROM AUTHOR]

This paper presents an analysis on the convergence to a steady state for the traffic dynamics in a network of signalized intersections. The links carrying the traffic from one intersection to another in the network are partitioned into a series of cells whose dynamics is modelled as a nonlinear cell transmission model. A microscopic convergence analysis is presented for the network where the traffic lights at the intersections are controlled by pre-timed control in which the average flow rate along each link is assumed to be known accurately. A macroscopic convergence analysis is applied to the traffic network where the traffic lights are controlled by coordinated model predictive control. In coordinated model predictive control, each local agent optimizes the switching times of traffic light at one intersection online with sampled local traffic state around this intersection and exchanged information from neighbouring intersections. A unique periodic function for the microscopic steady state is proved to exist under pre-timed control while a unique macroscopic steady state is achieved under coordinated model predictive control for the traffic lights. This paper validates by simulations that the ability of coordinated model predictive control to stabilize the traffic demand is not worse than pre-timed control even though the average flow rate along each link is not assumed to be known and further no explicit cycle time for the traffic light is suggested. Moreover, the incurred delay from the steady state is lower on the average under coordinated model predictive control compared to pre-timed control. • Presenting microscopic convergence analysis on pre-timed control • Proposing distributed model predictive control for traffic signals. • Presenting macroscopic convergence analysis on distributed control. • Validating claims on convergence of traffic signal control by simulations. [ABSTRACT FROM AUTHOR]

In the domain of complex urban traffic networks, real-time Traffic Signal Control (TSC) serves as a pivotal strategy for mitigating congestion. Coordinating signal control across multiple intersections involves considerable complexity. Deep Reinforcement Learning (DRL) has emerged as a robust solution. In recent years, there has been rapid advancement in TSC methods, with numerous researchers employing various novel DRL methodologies. Yet, existing surveys lack timeliness and universality in capturing the latest research. There is a notable gap in current research surveys with respect to the latest developments in TSC. Therefore, the focus of this paper lies in analyzing the most recent papers from the past five years, with the aim to provide a comprehensive and multi-dimensional review of the evolution of DRL in TSC. The survey categorizes current research based on model setups, utilized algorithms, and application scenarios. Finally, this paper highlights potential directions for future TSC research. [ABSTRACT FROM AUTHOR]

Urban traffic control (UTC) is inherently multiscale in both temporal and spatial domains. With the wide deployment of connected and automated vehicles (CAVs), data is increasingly available to help reveal this multiscale nature of UTC and the inter-dynamics among different scales. This paper applies the multiscale UTC framework we proposed earlier and extends it to UTC on a network of traffic signals with a mixed flow of CAVs and human-driven vehicles (HDVs). We adopt distributed control as the basic scheme for network-wide control and use information sharing to achieve cooperation among different intersections. We use CAV information to estimate HDV state and develop a "safety check" technique to control CAVs in the mixed traffic flow. Together, we propose a network-wide UTC framework with mixed traffic flow. To address the computation issue of the model-based multiscale method, we develop an imitation learning (IL) enhanced data-driven method to improve the computation efficiency. Specifically, we use a convolutional neural network (CNN) to represent the policies of the slower-scale signal control problem and use the data aggregation method as the learning framework to improve the policies. This algorithm's unique feature is that IL policies' training is based on the optimized results from the model-based multiscale control method. We test the model-based and IL-based methods in simulation, under various traffic scenarios and on multiple networks. We also test the transferability property of the IL-based method by training individual intersection control separately in small networks and applying them to larger networks with various types of intersections. [ABSTRACT FROM AUTHOR]

Traditional traffic signal control is one of the most widely used urban traffic control methods, due to its high efficiency, robustness, and simplicity. However, in the era of connected and automated vehicles (CAVs), it is prospective to enhance traffic control precision and achieve vehicle coordination in a more efficient manner. Focusing on an isolated intersection, this study proposes an innovative concept of Fine-grained Phase (FP) for CAV-enabled traffic control, which realizes more refined right-of-way allocation than traditional signal phases while preserving the simplicity and scalability. The development of FP jointly optimizes traffic control strategy and intersection layout design to fully utilize time-space resources, and multiple forms of FP could be generated in response to various traffic demands; meanwhile, vehicle coordination is achieved in an offline–online manner to balance solution quality and efficiency. Under the FP-based control, each lane is assigned with periodical entry time points using FP design results, then vehicles are supposed to catch up with the appropriate time points to fulfill their trips. Theoretical analyses are conducted under stationary vehicle arrival patterns, where the average delay and the intersection capacity associated with different forms of FP can be analytically acquired. For the implementation of FP-based control, we construct a general offline–online framework, consisting of an offline model that optimizes the geometry of the conflict zone at intersections and the control strategy of vehicles simultaneously, which is formulated as a mixed-integer linear programming problem, and an online FP adoption mechanism that can adapt to varying demand patterns. Under the framework, limited online computation is required, facilitating its practical real-time implementation. Simulation studies under a variety of demand settings show that FP-based control is advantageous over benchmark methods in reducing vehicle delay and also improving intersection capacity in most cases; even for cases with extremely imbalanced demands, FP-based control can admit comparable throughput, while the average delay is much smaller. In addition, results of computation time reveal that FP-based control could realize real-time implementation with negligible computation burden. • An innovative concept called FP for intersection traffic control with CAVs. • Joint optimization of traffic control and intersection layout design. • Offline–online implementation framework to balance solution quality and efficiency. • Theoretical analysis for average vehicle delay and admissible traffic capacity. • Improvement over signalized control and the state-of-art reservation strategy. [ABSTRACT FROM AUTHOR]

As our cities become more complex and traffic demand grows, managing such traffic efficiently becomes challenging. Hence, solutions that allow building upon the current traffic light systems and that can be readily deployed are of global interest. In this work, we address the challenge of improving traffic light management at intersections. We propose an agent-based traffic light control system where an agent, one per intersection, dynamically regulates the light's phase cycle depending on the current traffic conditions. To this end, we will rely on Deep Networks to adequately train agents to make good decisions. Simulation results in a realistic scenario using SUMO show that our proposed approach can significantly reduce waiting times, improving transit times by 44% compared to the standard fixed-timing method. Additionally, to assess the effectiveness and reliability of our control algorithm, we introduce new performance metrics. [Display omitted] • A distributed neural network-based agent system optimizes traffic in superblocks. • Every agent manages one intersection using deep Q-learning with experience replay. • Each agent optimizes its intersection and reduces vehicle waiting times. • Compared with fixed time, our algorithm achieves a 44% reduction in waiting times. • Thanks to deep Q-networks, changes in agents' actions allow traffic light adaptation. [ABSTRACT FROM AUTHOR]

The efficiency of urban traffic management and congestion alleviation relies heavily on accurate forecasting of Origin-Destination (O-D) demand matrices. Existing models primarily focus on estimating O-D demand for various travel purposes throughout the day, which is characterised by its pulsating nature. However, these models often compromise the precision of peak-hour forecasts, leading to unreliable dynamic traffic control and challenges in effectively reducing peak-hour congestion. To tackle this challenge, this paper proposes a novel method for predicting commuting O-D demand matrices. Our method employs community detection algorithms on road networks to precisely partition commute O-D regions, incorporating Points of Interest (POIs). We also present a spatio-temporal dynamic weighted hypergraph model that leverages these partitioned regions, time characteristics from observed O-D trips, and meteorological data to improve forecasting. Comparative analyses with contemporary models and ablation studies indicate our method significantly enhances prediction accuracy, by approximately 5%. These findings imply that the proposed method more effectively encompasses the varied characteristics of commuting during peak hours, thereby providing more accurate demand matrices for urban traffic management. • An integrated framework for community detection and hypergraph representation. • Peak-hour urban commute patterns are captured with embedded POI features. • Time segment and weather condition are used to enhance prediction accuracy. • A spatio-temporal dynamic weighted hypergraph model for O-D prediction. [ABSTRACT FROM AUTHOR]

Large-scale congestion can lead to traffic paralysis, which severely hampers the flow of vehicles and disrupts the normal functioning of urban traffic. Traffic optimization strategies can effectively improve the performance of road networks, but often ignore the impact of regional traffic conditions and equity. This paper presents a novel traffic strategy to solve regional traffic congestion in large cities, particularly focusing on mixed traffic scenarios of connected and non-connected vehicles. The proposed method involves monitoring the traffic condition of the congestion warning community and adjusting the internal access flow within each region. The problem is formulated as a Stackelberg game, with traffic policymakers and road users as the key players. The upper layer aims to control traffic access by issuing a community warning index, with the objective of minimizing congestion warning conditions within the community. This information is then disseminated to connected vehicles which utilize it to generate personalized route guidance, while non-connected vehicles remain unaffected. The lower-level objective is to allocate vehicles in the transportation network in a user-optimal manner. To solve the bi-level programming model, the paper introduces a variable neighborhood search approach based on graph theory. The Frank-Wolfe algorithm is used to solve the lower-level model, with a penalty function introduced to transform the constrained traffic assignment problem (TAP) into an unconstrained TAP. The proposed method is applied using the data of Beijing urban road network and a sensitivity analysis is conducted to examine the impacts of critical parameters, such as regional partitioning and mixed traffic proportion. The results show that the method exhibits improved optimization performance across different parameter settings, effectively utilizing idle links and contributing to a reduction in the occurrence of traffic warning regions. • The regional control strategy prevents the emergence of warning communities. • A bi-level model is developed to solve the traffic diversion of mixed traffic. • A variable neighborhood search algorithm is designed to solve the model. • The control strategies under different critical variables scenarios are analyzed. [ABSTRACT FROM AUTHOR]

Urban traffic management and traffic signal control systems, denoted as Urban Traffic Control (UTC) systems, are used extensively worldwide by Local Government Authorities (LGAs) when implementing strategic transport policies. However, it is not clear how well the requirements imposed by LGA policy implementation will be met by UTC systems being developed for the future. Therefore, research was undertaken to analyse how delivery of urban transport policies over the next 5–10 years would shape LGAs' requirements for the next generation of UTC systems, and thereby to identify Key Performance Indicators (KPIs) to provide user-led guidance for future system development. A two-stage survey of LGA policy makers and implementers from around the world was conducted. The results produced consensus among the group of participants (n = 16) on 17 KPIs, representing a synthesis of expert opinions on the desired features of future UTC systems from a policy implementation perspective. The research makes an important contribution in eliciting the wide-ranging breadth of issues associated with delivering strategic transport policies and understanding how these issues affect the requirements LGAs have for future UTC systems. Aligning future UTC system capabilities with LGAs' requirements will enable more effective implementation of strategic urban transport policies worldwide and allow the benefits to society associated with those policies to be realised. • Local Government Authorities (LGAs) implement strategic transport policies. • Policy implementation is achieved using urban traffic management systems. • LGA policy makers and implementers were interviewed about desired system features. • Consensus was achieved on Key Performance Indicators (KPIs) for future systems. • The KPIs provide user-led guidance to align system capabilities with LGA policies. [ABSTRACT FROM AUTHOR]

• Urban traffic control is achieved by optimal route guidance based on avoidance maps. • Avoidance maps define which areas vehicles should better circumvent to sustain optimal internal traffic conditions. • Avoidance maps are derived from a reactive proportional control scheme. • A thorough sensitivity analysis shows how to tune the main parameters of the new concept (region sizes, safety distance, controller time horizon). • The new concept can improve total travel time by around 15% with little harm to individual vehicles. • Privacy by design is guaranteed as vehicles are not sharing data with the control system. This paper presents a new concept for operating traffic management at a sizeable urban scale. The overall principle is to partition the network into multiple regions, where traffic conditions should remain optimal. Instead of monitoring the inflow, like for perimeter control, deviations in regional mean speed compared to a reference value are transformed into avoidance levels by a centralized controller. That information is broadcasted to vehicles through a public avoidance map. The onboard navigation systems then interpret this map to deliver individual route guidance according to the avoidance levels. In that way, the concept tends to distribute vehicles in the network optimally while preserving privacy. In addition to the controller parameters itself, three other critical parameters define the system: the safety distance, the controller time horizon, and the region sizes. Thorough sensitivity analysis leads to the optimal setting. The concept is proven effective using microsimulation for both a Manhattan and a realistic network. The total travel time is improved by about 15% when traffic is severely congested compared to the uncontrolled case. In the meantime, the mean individual travel distance increase for rerouted vehicles is kept below 10%. Those results have been obtained with a simple decentralized reactive control framework, i.e., an individual proportional feedback system independently governs each region. More advanced formulations introducing cooperation between the regions are also tested. In a nutshell, this paper is a proof of concept for a new control system that appears both practical and valuable to alleviate congestion in urban areas. © 2021 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 24th International Symposium on Transportation and Traffic Theory [ABSTRACT FROM AUTHOR]

Intelligent Transportation Systems (ITS) face the formidable challenge of large-scale missing data, particularly in the imputation of traffic data. Existing studies have mainly relied on modeling network-level spatiotemporal correlations to address this issue. However, these methods often overlook the rich semantic information (e.g., road infrastructure, sensor location, etc.) inherent in road networks when capturing network-wide spatiotemporal correlations. We address this limitation by presenting the Graph Transformer-based Traffic Data Imputation (GT-TDI) model, which imputes missing values in extensive traffic data by leveraging spatiotemporal semantic understanding of road networks. The proposed model leverages semantic descriptions that capture the spatial and temporal dynamics of traffic across road networks, enhancing its capacity to infer comprehensive spatiotemporal relationships. Moreover, to augment the model's capabilities, we employ a Large Language Model (LLM) and prompt engineering to enable natural and intuitive interactions with the traffic data imputation system, allowing users to query and request in plain language, without requiring expert knowledge or complex mathematical models. The proposed model, GT-TDI, utilizes Graph Neural Networks (GNN) and Transformer architectures to perform large-scale traffic data imputation using deficient observations, sensor social connectivity, and semantic descriptions as inputs. We evaluate the GT-TDI model on the PeMS freeway dataset and benchmark it against cutting-edge models. The experimental evidence demonstrates that GT-TDI surpasses the cutting-edge approaches in scenarios with intricate patterns and varying rates of missing data. • A novel semantic description for traffic data imputation that captures spatiotemporal correlations flexibly. • A pioneering application of prompt engineering to ITS that bridges the gap between LLMs and ITS. • The proposed model leverages LLMs and semantic prompts for traffic data imputation. • Extensive experiments on real-world traffic data show the superiority. • A potential framework for integrating LLMs and semantic prompts into other ITS tasks. [ABSTRACT FROM AUTHOR]

The store-and-forward traffic model is widely used because of its simplicity and reasonable comprehension. Its results determine the duration of the green light according to a preset cycle. This paper builds further on the currently existing model: it expands the definitional domain of the optimization problem by adding additional probabilistic requirements regarding the number of vehicles in a street segment between intersections controlled by traffic lights. The defined optimization problem minimizes the probability of increasing the number of vehicles in the considered transport network. A further extension of the store-and-forward model allows the control domain to contain both green lights and cycle durations. Such optimal control allows the achievement of additional objective functions such as minimizing the number of vehicles in the network and minimizing the total time spent through the network. The definition of this control problem is done through a bi-level hierarchical formalization involving the store-and-forward model. The derived extended models are numerically tested and compared with real cases, giving preference to the derived extensions of the classical store-and-forward model. [ABSTRACT FROM AUTHOR]

• Fusion process between signals shows that traffic flow undergoes phase transition. • Traffic network is modelled as an Ising model with traffic flow taking on two states. • Transition from congestion becomes more apparent when control undergoes Parrondo switching. While increasing urban traffic can be an indicator of development, this inevitably results in traffic congestion in urban road networks. Is there a way to manage traffic flow through the control of traffic signals such that the overall network congestion is improved? Traffic light signals can be represented as two states of an Ising model. It is possible for traffic lights to "communicate" with each other through a fusion process from a remote management control system. This requires collection of information which can be fed to a centralized decision-making control mechanism. We first explore the fusion process between traffic signals and show that it is possible for traffic flow in a city to follow the phase transition as exhibited in the 2D Ising model. The model will be extended to show that a random switching between signalling control mechanisms can result in congested traffic being susceptible to transit out of congestion. [ABSTRACT FROM AUTHOR]

• Centralised and decentralised algorithms with optimal control formulation. • Sensitivity based decentralised signal optimisation algorithms. • Comparison with distributed backpressure control policy. • Investigation of different network decomposition schemes. • Case study of optimisation algorithms on Central London networks. This study develops and analyses the centralised and decentralised solution procedures for urban network traffic management through an optimal control framework. The optimal control is formulated based upon the Hamilton-Jacobi formulation of kinematic wave model. Following the problem decomposition under the decentralised scheme, we introduce the use of semi-analytical performance derivatives when developing the decentralised solution algorithm. The proposed control strategies are applied to a set of test scenarios constructed from a real road network in Central London in the UK. Specific interests in the investigation include comparison of the performance gain and computational effort of the two strategies under different circumstances. We also investigate effect of network decomposition strategies on the performance of the solution algorithm. This study generates insight on urban traffic management with use of traffic flow theory, decentralised optimisation, and network decomposition techniques. [ABSTRACT FROM AUTHOR]

• Designing a macroscopic modeling framework for perimeter gating control and dynamic user equilibrium. • Designing a routing scheme with instantaneous dynamic user equilibrium and predictive travel time. • Application of Proportional-Integral (PI) controller for traffic control. • Investigating the impact of perimeter gating control on routing inside-and-outside the reservoir. • Analysis of emissions in the traffic network. In recent years, several perimeter control strategies have been proposed for traffic management in cities. The common factor found in these works is the use of Macroscopic Fundamental Diagram (MFD) models to describe the dynamics of the network and optimize traffic inside the perimeter by manipulating perimeter inflows. Perimeter gating control strategies are attractive for traffic management inside the inner city. However, it inevitably creates a negative impact on the traffic outside. Most of the works in this research area have neglected vehicle re-routing outside the controlled perimeter, i.e., they do not consider demand elasticity to the central region resulting from gating and the related queues. In this paper, we propose a global modeling framework capable of assessing the effect of perimeter gating control (in terms of queue, emission, and total time spent) on the full network, considering demand elasticity resulting from Dynamic User Equilibrium (DUE). Classical Proportional-Integral (PI) control scheme is used to control traffic congestion inside a central region (reservoir). The modeling framework is comprised of: (i) an accumulation-based MFD model to reproduce traffic dynamics inside the reservoir, (ii) point-queue model to represent queuing vehicles on inbound links to the gating points, and (iii) a time-dependent travel time profile based on a steady-state approximation of MFD dynamics to characterize the alternative road network (bypass). DUE is then implemented, considering instantaneous predicted travel time. This determines how the demand to the inner region is affected by the gating. The functioning of the global system is assessed by total time spent and NOx and CO 2 emissions inside the reservoir and for the full network. The presented simulation results show that the perimeter gating control helps to maintain congestion at the desired level with significant improvements in the total time spent and the mean speed in the network. However, it shows a slight increase in the queues. As expected, deviation to the bypass alternative is significant and should not be neglected when carrying out a global assessment of gating system performance. [ABSTRACT FROM AUTHOR]

The perimeter control concept based on macroscopic fundamental diagram (MFD) modeling is now well-established in various literature. Recent research efforts are devoted to develop perimeter control schemes, which can be eventually applied to urban traffic control systems. Modern urban traffic control systems increasingly rely on information technology infrastructure, combining various network and physical facilities and use of communication technologies, which increases the vulnerability for cyberattacks. As witnessed in recent real-life events, modern urban traffic control systems are vulnerable and not protected from internal and external, malicious and accidental threats. Hence, applicable perimeter control algorithms should be robust not only against dynamic uncertainties, but in addition, it should be resilient against cyberattack issues for real future implementations. In this paper, a resilient perimeter control scheme is developed for a modified multi-region MFD system model. The developed perimeter control is a fully decentralized scheme, which operates based on online local regional information only. It achieves an a priori desired tracking performance with simultaneous compensation of (i) cyberattacks in the channels of control signals transmission, (ii) model parameter uncertainties, (iii) unknown but bounded disturbances, and (iv) control input constraints. A framework of adaptive control techniques is utilized to design the perimeter control. The proposed fault-tolerant perimeter control does not require on-line fault detection, as the fault effects caused by cyberattacks are continuously and adaptively compensated by the developed algorithm. Stability proof of the closed-loop system in terms of the tracking and parameter errors are introduced, and simulation results for three aggregate urban traffic regions are presented. [ABSTRACT FROM AUTHOR]

Active traffic management is the ability to dynamically manage congestion based on prevailing and predicted traffic conditions. Several technologies and algorithms have been employed to develop a set of tools utilised by Traffic Control/Traffic Management Centres with the aim to mitigate disruptions and improve traffic flow in the road networks. One such tool, belonging to the Advanced Traveller Information Systems category is Variable Message Signs. Variable Message Signs are located in major roads and provide real-time traffic information to the drivers so as to modify their behaviour considering the prevailing traffic conditions. To this end, the question arises. Is active traffic management in urban road networks – as currently provided by Variable Message Signs – effective for professional drivers? In plain words, do professional drivers modify their route choice, i.e. divert under the provision of real-time traffic information? A stated preference survey was performed to truck and taxi drivers to identify their response to information provided by Variable Message Signs. Results, in the form of ordered probit models indicated that the provided information is not that relevant to the investigated driver categories, as it is for passenger car drivers. [ABSTRACT FROM AUTHOR]

• We propose a new, max-pressure based network traffic control policy for legacy and autonomous vehicles. • We introduce an autonomous vehicle-restricted phase, named the blue phase, to coordinate autonomous vehicles at network intersections. • We formulate a green signal phase model wherein movement capacities are endogenous to the choice of the activation matrix and lane FIFO blocking effects are accounted for. • We show that the proposed hybrid network traffic control policy is stable, i.e. maximizes throughout. • Our numerical experiments reveal considerable trade-offs in terms of vehicle-class travel time based on the level of congestion and the proportion of autonomous vehicles in the network. With the forecasted emergence of autonomous vehicles in urban traffic networks, new control policies are needed to leverage their potential for reducing congestion. While several efforts have studied the fully autonomous traffic control problem, there is a lack of models addressing the more imminent transitional stage wherein legacy and autonomous vehicles share the urban infrastructure. We address this gap by introducing a new policy for stochastic network traffic control involving both classes of vehicles. We conjecture that network links will have dedicated lanes for autonomous vehicles which provide access to traffic intersections and combine traditional green signal phases with autonomous vehicle-restricted signal phases named blue phases. We propose a new pressure-based, decentralized, hybrid network control policy that activates selected movements at intersections based on the solution of mixed-integer linear programs. We prove that the proposed policy is stable, i.e. maximizes network throughput, under conventional travel demand conditions. We conduct numerical experiments to test the proposed policy under varying proportions of autonomous vehicles. Our experiments reveal that considerable trade-offs exist in terms of vehicle-class travel time based on the level of market penetration of autonomous vehicles. Further, we find that the proposed hybrid network control policy improves on traditional green phase traffic signal control for high levels of congestion, thus helping in quantifying the potential benefits of autonomous vehicles in urban networks. [ABSTRACT FROM AUTHOR]

• This paper proposes a network-wide method to identify intersection traffic congestion in a road network. • The proposed method is solely based on low-frequency probe vehicle data, i.e., map-independent. • It is able to rapidly and approximately detect intersection congestion along all turning directions. • It is a computer-aided method in the era of big data can greatly decrease traffic engineers' workload. Locating the bottlenecks in cities where traffic congestion usually occurs is essential prior to solving congestion problems. Therefore, this paper proposes a low-frequency probe vehicle data (PVD)-based method to identify turn-level intersection traffic congestion in an urban road network. This method initially divides an urban area into meter-scale square cells and maps PVD into those cells and then identifies the cells that correspond to road intersections by taking advantage of the fixed-location stop-and-go characteristics of traffic passing through intersections. With those rasterized road intersections, the proposed method recognizes probe vehicles' turning directions and provides preliminary analysis of traffic conditions at all turning directions. The proposed method is map-independent (i.e., no digital map is needed) and computationally efficient and is able to rapidly screen most of the intersections for turn-level congestion in a road network. Thereby, this method is expected to greatly decrease traffic engineers' workloads by providing information regarding where and when to investigate and solve traffic congestion problems. [ABSTRACT FROM AUTHOR]

Recently, with increasing urban traffic congestion, there has been an upsurge in studies on reinforcement learning for traffic signal control (RL-TSC), which enables efficient traffic management. However, most existing RL-TSC research relies on simulation, and instances of real-world application are relatively scarce. Furthermore, existing RL-TSC methods employ a step-based approach, controlling traffic signals every short step. This approach can be inefficient in oversaturated traffic conditions, where large deviations in traffic demand between movements prevent traffic signals from being effectively managed based on the overall traffic situation. In this study, we aim to transform simulation-based RL-TSC into a more practical and applicable model. We have developed a RL-TSC method capable of responding to oversaturated traffic conditions, designing an action space that enables the agent to derive an optimal signal set for every cycle length, thereby understanding the traffic situation across all movements. During each cycle length, our proposed model conducts traffic signal optimization and identifies the optimal signal through an iterative exploration. To facilitate a fast and accurate strategy search process, we developed a kinematic wave-based mesoscopic model. This model estimates the density across the entire link, based on collected traffic data and derives the state and reward values. We have validated the field applicability of the proposed RL-TSC method through a real-world demonstration at a congested intersection in Seoul, Korea. The results indicated a significant improvement in traffic congestion, with the average queue length at the intersection reduced by up to 11.4%. [ABSTRACT FROM AUTHOR]

Controlling traffic in large-scale urban road networks is a challenging problem. Aggregated dynamical models, based on the macroscopic fundamental diagram (MFD) of urban traffic, enable model-based control design. As an alternative to perimeter control actuation commonly used in MFD-based control, in this paper, we propose actuation over regional space-mean speeds, which we name average speed control. The method involves manipulation of regional speeds via instrumentation similar to variable speed limits in freeways, or using vehicle-to-infrastructure communication. We develop nonlinear model predictive control schemes considering actuation over average speed and perimeter control. Their performances are compared using simulations on congested scenarios, the results of which suggest potential of the method as an alternative or complementary actuation to perimeter control. • A novel speed control strategy for urban networks. • Manipulation of regional average speeds to optimize traffic conditions in the whole network. • An MPC scheme combining perimeter control and average speed control. [ABSTRACT FROM AUTHOR]

Model predictive control (MPC) has been widely used for traffic management, such as for minimizing the total time spent or the total emissions of vehicles. When long-term green urban mobility is considered including e.g. a constraint on the total yearly emissions, the optimization horizon of the MPC problem is significantly larger than the control sampling time, and thus the number of the variables that should be optimized per control time step becomes very large. For systems with dynamics that involve nonlinear, non-convex, and non-smooth functions, including urban traffic networks, this results in optimization problems that are computationally intractable in real time. In this paper, we propose a novel bi-level temporal distribution of such complex MPC optimization problems, and we develop two mathematically linked short-term and long-term MPC formulations with small and large control sampling times that will be solved together instead of the original complex optimization problem. The resulting bi-level control architecture is used to solve the two MPC formulations online for real-time control of urban traffic networks with the objective of long-term green mobility. In order to assess the performance of the bi-level control architecture, we perform a case study where a rough version of the model of the urban traffic flow, S-model, is used by the long-term MPC level to estimate the states of the urban traffic networks, and a detailed version of the model is used by the short-term MPC level. The results of the simulations prove the effectiveness (with respect to the objective of control, as well as computational efficiency) of the proposed bi-level MPC approach, compared to state-of-the-art control approaches. • A bi-level MPC-based architecture is implemented for green mobility. • The link of the control levels is based on the long and short-term allowed emissions. • This framework can limit the emissions with an affordable online computation time. [ABSTRACT FROM AUTHOR]

Intelligent transport systems (ITS) are going to dominate the landscape of traffic control in the foreseeable future. In order to meet the travel requirements of individuals, ITS can efficiently use the available traffic infrastructure, guarantee the protection of transportation and urban traffic, facilitate public safety monitoring, and provide efficient control mechanisms. The term "Multi-criterion Group Decision-Making" (MCGDM) refers to a circumstance in which numerous criteria must be considered before making a decision. Hence, it is important to develop novel Non-linear-Diophantine fuzzy Bonferroni-mean operators to effectively find the optimal alternative based on specific criteria for Intelligent Transportation Systems. Numerous non-linear-Diophantine fuzzy Bonferroni-mean-Dombi Aggregation-Operators are introduced (Non-linear-Diophantine Fuzzy-Weighted-Averaging (N-LDFWA) operators and Non-linear-Diophantine Fuzzy-Weighted-Geomatric (N-LDFWG) operators), employing the Bonferroni-Mean-Dombi T-norm and T-conorm. In order to examine the reliability and efficiency of the present method, a real-life scenario of ITS is used. The execution performed for the evaluation of the intelligent transport system and some comparisons are also provided. [Display omitted] • In the context of Bonferroni Mean Dombi, to propose a new nonlinear Diophantine fuzzy set. • Establishing six new operators for NLDFNs' for selection criteria. • Improving intelligent transportation systems can provide a competitive advantage to enterprises. [ABSTRACT FROM AUTHOR]

Traffic congestion has been regarded as one of the most challenging problems in the urban transportation systems. The problem becomes trickier at traffic bottlenecks where the heavy traffic flow spills over and over-saturation appears, and the coordination of traffic control strategies in a network scale is required. With the information availability and control precision enabled by connected and automated vehicle (CAV) technologies, this study proposes a network-level traffic coordination method with dynamic entrance holding in a mixed CAV traffic, and establishes a two-layer framework to develop traffic control strategies to handle the over-saturation problem. In the upper layer, the coordination of signal control among intersections is optimized to minimize travel cost within the network. The upper bound of travel delays under the control strategy is theoretically derived and the critical traffic volume is specified for designing the entrance holding strategy. To accelerate the solving process, a two-stage solution method along with a series of valid inequations are developed, and the feasibility and optimality of the solution are investigated. In the lower layer, a dynamic entrance holding strategy is put forward to regulate the entering traffic at network perimeters to avoid over-saturation and improve traffic orderliness. To enhance the holding effectiveness, a traffic prediction method is created to capture the future traffic evolution based on the real-time information, and combined with the flow reservation mechanism to proactively adjust the holding strategy in real time. Simulation experiments are conducted to evaluate the performance of the proposed methodologies, and compare with other control strategies. The results show that the coordinated method could substantially improve traffic mobility within the network, especially in high traffic demands. • A joint control scheme combining coordinated TSC strategy and entrance hold strategy • A network-level TSC optimization model for minimizing vehicular delays • A two-stage solution method with valid inequations to improve solving efficiency • A dynamic entrance holding strategy by integrating flow reservation mechanism • Simulation experiments to validate the performance of the proposed methodologies [ABSTRACT FROM AUTHOR]

Accurate and real-time traffic forecasting is of great significance for urban traffic planning, traffic control, and traffic management. However, the time-varying dynamic spatial relations and the complicated spatial–temporal dependencies are still open problems to be considered in traffic forecasting. To address these issues, we propose a graph convolutional recurrent network for traffic forecasting with a dynamic adjacency matrix within an encoder–decoder framework, named DAGCRN. The DAGCRN consists of a spatial relation extraction module (SREM), an adjacency matrix update module (AMUM), a dynamic graph convolutional recurrent module (DGCRM), and a global temporal attention module (GTAM). Specifically, SREM and AMUM are proposed to capture nodes' mutual relations at each time step and to model the evolution of the dynamic adjacency matrix, respectively. DGCRM captures the spatial–temporal dependencies of traffic data based on dynamic graph convolution and gated recurrent unit. GTAM is designed to extract the long-range temporal dependencies between future time steps and historical time steps. Extensive experiments on two real-world traffic speed datasets demonstrate that the proposed DAGCRN outperforms a number of representative baselines consistently. [ABSTRACT FROM AUTHOR]

Speed limit policies are commonly adopted to manage and control traffic in urban areas due to their effectiveness and ease of implementation. Comprehending the complete effect of a speed limit policy is complicated and requires modeling and quantified investigations. In this paper, we propose a comprehensive simulation-based framework to assess the potential implications of different speed limit policies in urban residential areas. The framework models the policy impacts related to road safety (risk exposure for pedestrians and driving safety), traffic efficiency (travel time) and the environment (fuel consumption, exhaust emissions and noise exposure), using microscopic traffic simulation. The evaluations are conducted at various spatial granularity levels, i.e., link level, route level, origin–destination (OD) level and network level, and can be further utilized to develop relationship models between the key performance indicators (KPIs) and simulation inputs. The framework is implemented in an urban area located in the city center of Munich, Germany, and multiple speed limit scenarios are designed and compared. The results show that speed limit reduction can significantly improve road safety and environmental externalities within the modeled network/area with a relatively small cost to traffic efficiency. Such a framework can be used as an economical evidence collection method for an evidence-based policymaking approach to speed limit policies. The proposed simulation-based framework can also be further extended to adapt the assessment of other traffic-related policies. [ABSTRACT FROM AUTHOR]

• Systematic review on urban traffic control under mixed traffic with CAVs. • Cover signal timing, CAV trajectory/route planning, and their cooperative control. • Summarize mixed traffic control methods at intersections with CAV-dedicated facilities. • Outline promising research directions and challenges. Efficient traffic control can alleviate traffic congestion, reduce fuel consumption, and improve traffic safety. With the development of communication and automation technologies, regular vehicles (RVs), connected vehicles (CVs), and connected and automated vehicles (CAVs) will coexist on urban roads in the near future. In this paper, we systematically review the studies on urban traffic control under the mixed traffic environment, which utilize CVs and CAVs to improve the performance of urban traffic control. We first give a glance at the evolution of urban traffic control with the deployment of CVs and CAVs to identify the important and promising research fields in the mixed traffic control. We then review the recent advances in traffic signal timing with CAV/CV trajectory data, CAV trajectory/route planning, and the cooperative control of CAV trajectories and traffic signals from the intersection level to the network level under the mixed traffic environment. Besides, we summarize two types of studies on mixed traffic control at intersections with CAV-dedicated facilities: 1) traffic signal timing and vehicle trajectory planning at intersections with CAV-dedicated lanes, and 2) deployment of CAV-dedicated facilities including CAV-dedicated lanes, links, and zones. Finally, the future research directions and challenges are discussed. We hope that this review could provide researchers with a helpful roadmap for future research on urban traffic control under the mixed traffic environment. [ABSTRACT FROM AUTHOR]

Urban traffic control is multi-scale in both temporal and spatial domains, which is of great importance to improve the quality and efficiency of our transportation system. In this paper, we propose a general multi-scale modeling and control framework for urban traffic control with connected and automated vehicles (CAVs). We apply the proposed multi-scale framework to explicitly model and solve the two-scale signal-vehicle coupled control (SVCC) problem under the environment of full penetration of CAVs. To solve the SVCC problem, we propose a model predictive control (MPC) scheme and develop a stability analysis method based on the concept of consistency of key states between the two scales. We show numerical results of the SVCC model and comparisons with some benchmark methods. We also discuss how to extend the SVCC model to the mixed traffic flow and tri-scale/multi-scale urban traffic control problems. • Urban traffic control (UTC) is multi-scale in both temporal and spatial domains. • A general multi-scale control framework is developed for UTC with CAVs. • The proposed framework is applied to two-scale signal-vehicle coupled control (SVCC). • A model predictive control (MPC) scheme is developed to solve the SVCC model. • Stability analysis is conducted based on the consistency of key states of the two scales. [ABSTRACT FROM AUTHOR]

• Propose priority-based infrastructure and control aware user heterogeneity, i.e. value-of-time. • Develop a joint optimization framework for traffic signal control and priority-based pricing. • Utilize connected vehicles to inform traffic flow and value-of-time information. • Test the proposed joint framework in real-city settings that resemble Zurich network. • Demonstrate traffic performance can be significantly improved while equitable savings achieved. Real-time control of large-scale urban networks has been attracting significant research attention. This paper, using the information provided by connected vehicles, proposes a novel control approach based on the concept of perimeter control to maximize the social welfare of all vehicles. The contributions of this paper are threefold. First, we consider vehicle heterogeneity (i.e. corresponding to different transportation modes, or with different occupancies, values of time, priority levels, etc.) and integrate a priority scheme into perimeter control to improve both the traffic performance and the social welfare. This is achieved by installing priority lanes at some of the perimeter intersections. Unlike the existing research works that provide priority to certain traffic modes, we dynamically identify the groups of vehicles that we should prioritize, in order to maximize the social welfare. Second, we develop a model predictive control approach to simultaneously optimize the toll for using the priority lanes and the traffic signal timings at the perimeter intersections. This approach can explicitly handle the constraint of the storage capacity of each intersection link. Third, we propose a recursive estimation algorithm to update our knowledge on the distribution of the value of times (VOTs), using the lane choice information of connected vehicles. The proposed approach is tested in a simulated network which resembles the main features of the city center of Zurich, Switzerland. By using the proposed strategy, the traffic accumulation inside the network is still stabilized, and the monetary costs due to delay are significantly reduced at the entire network (up to 25.8%) compared to the strategy without priority. The distribution of the combined cost (including cost due to delay and tolls) is more uniform across VOT groups than that resulting from the strategy without priority. It is also shown that the proposed recursive estimation algorithm quickly converges and further improves the social welfare. [ABSTRACT FROM AUTHOR]

• Development of an integrated optimization program for cooperative signal timing and perimeter control. • Development of a distributed methodology to find solutions to the cooperative problem in real-time. • Increasing network throughput by 6.0–12.8% compared to separate signal and perimeter control. This paper presents an integrated formulation and a distributed solution technique for cooperative signal control and perimeter traffic metering in urban street networks with various market penetration rates of connected vehicles. The problem is formulated as a mixed integer nonlinear program thus, does not scale well with the size of the network in a centralized optimization framework due to the presence of many mixed integer decision variables and nonlinear constraints. To address this limitation, we will develop a distributed model predictive control that distributes the network-level cooperative problem into several intersection-level sub-problems and coordinates their decisions. Our numerical analyses show that the proposed distributed methodology finds solutions to the problem in real-time with the optimality gap of at most 3.6% in our case studies. We have implemented the distributed methodology in Vissim and observed that cooperative signal timing and perimeter control yielded significant improvements in traffic operations. Our case study results show that the cooperative approach increases the number of completed trips by 6.0–12.8% and 10.9–11.0% and reduces the total travel times by 8.1–9.0% and 23.6–24.2% compared to independent signal control and independent perimeter control, respectively. [ABSTRACT FROM AUTHOR]

Abstract Network-wide traffic control strategies (e.g. perimeter control and route guidance) in urban networks have recently been mainly studied to relieve or postpone congestion based on the theory of macroscopic fundamental diagram (MFD). Nevertheless, these studies are mostly applied to the statically partitioned networks or the dynamic networks that fail to fully consider traffic state prediction, conflicting with strongly spatiotemporal variability of traffic congestion or objective of active traffic management (ATM). This paper proposes a methodology to dynamically identify critical congestion warning areas from heterogeneous urban road networks, which aids to design efficient perimeter control approaches. In the methodology, a dynamic directed weighted network is built on the base of the link connectivity and the real-time traffic loads, and a link travel time prediction method based on Kalman filter is developed to calibrate directed weight values and undirected input values for links. With the undirected link input information, a dynamic congestion warning community detection method which consists of three consecutive steps is developed. Firstly, it could capture emergence of new congestion areas based on the definition of congestion seed intersection. Secondly, expansion and regression of each congested area could be achieved with the objectives of spatial compactness and traffic condition homogeneity. Thirdly, a two-level merging algorithm of adjacent different congestion areas is designed utilizing modularity model in community detection of complex networks. The proposed methodology is validated using ground truth data from downtown network of Jinan City in China. The results show that the proposed algorithms can efficiently track congestion evolutionary processes and effectively detect congestion warning areas from the test real network. Highlights • First, we use community detection theory to identify traffic congestion warning areas. • Second, three consecutive steps are proposed to describe the entire process of traffic congestion propagation. • Third, homogeneous sub-areas with different levels of congestion are extracted from heterogeneous urban network. • Fourth, our method performs more effectively in real-word urban road networks. [ABSTRACT FROM AUTHOR]

Highlights • Provide a systematic review of CAV based UTC studies in the last few decades. • Highlight the change of design philosophy for UTC. • Summarize the benefits and drawbacks of current CAV-based UTC methods. • Predict the future study directions. Abstract Inefficient traffic control is pervasive in modern urban areas, which would exaggerate traffic congestion as well as deteriorate mobility, fuel economy and safety. In this paper, we systematically review the potential solutions that take advantage of connected and automated vehicles (CAVs) to improve the control performances of urban signalized intersections. We review the methods and models to estimate traffic flow states and optimize traffic signal timing plans based on CAVs. We summarize six types of CAV-based traffic control methods and propose a conceptual mathematical framework that can be specified to each of six three types of methods by selecting different state variables, control inputs, and environment inputs. The benefits and drawbacks of various CAV-based control methods are explained, and future research directions are discussed. We hope that this review could provide readers with a helpful roadmap for future research on CAV-based urban traffic control and draw their attention to the most challenging problems in this important and promising field. [ABSTRACT FROM AUTHOR]

Highlights • The human decision making in car following tasks based on Wiedemann is introduced. • A driving simulator study is executed to verify these mechanisms. • At higher velocities, decisions tend to be made based on time headway. • This verifies the Wiedemann mechanism. • At lower velocities, decisions are also made based on time to collision. Abstract In the present paper, a theory-based advancement of the car following model of Wiedemann is introduced. It is suitable for a more realistic simulation of human longitudinal guidance behaviour in the lower velocity range, like in urban traffic. The algorithm was identified and developed by means of a driving simulator study, which is introduced as well. While the decision making process in recent car following models based on Wiedemann is solely determined by the time headway to the leading vehicle, the results of the study show that information about time to collision is also used in the lower velocity range. This approach delivers a more realistic simulation of human car following behaviour, which can be integrated into driver models for microscopic traffic simulations and for the control of surrounding traffic in driving simulators. [ABSTRACT FROM AUTHOR]

Abstract Electric bicycle (EB) has been one popular traffic tool in urban traffic system, but it has some disorder motion behaviors at a signalized intersection. In this paper, we propose a strategy to control each EB’s motion at a signalized intersection, and design some rules to explore each EB’s motion and the impacts of the proposed strategy on each EB’s motion at the signalized intersection. The numerical results indicate that the proposed strategy can enhance the operational efficiency at the signalized intersection. The results can help administrator to better manage the EB’s motion at the signalized intersection. Highlights • A strategy for EB flow is proposed. • EB’s motion rules under the proposed strategy are designed. • The effects of the proposed strategy on each EB’s motion studied. [ABSTRACT FROM AUTHOR]

Highlights • Apply CTM to urban traffic including traffic lights, queue discharge, merge/diverge. • Implement and validate proposed model by comparing to SUMO results, and explain differences. • Develop local MPC selecting switching times of traffic lights, at time resolution of a few seconds. • Evaluate performance of traffic system using local MPC control for traffic lights, for small urban region. Abstract This paper is the first in a series of reports presenting a framework for the hierarchical design of feedback controllers for traffic lights in urban networks. The goal of the research is to develop an easy to understand methodology for designing model based feedback controllers that use the current state estimate in order to select the next switching times of traffic lights. In this paper we introduce an extension of the cell transmission model that describes with sufficient accuracy the major causes of delay for urban traffic. We show that this model is computationally fast enough such that it can be used in a model predictive controller that decides for each intersection, taking into account the vehicle density as estimated along all links connected to the intersection, what switching time minimizes the local delay for all vehicles over a prediction horizon of a few minutes. The implementation of this local MPC only requires local online measurements and local model information (unlike the coordinated MPC, to be introduced in the next paper in this series, that takes into account interactions between neighbouring intersections). We study the performance of the proposed local MPC via simulation on a simple 4 by 4 Manhattan grid, comparing its delay with an efficiently tuned pretimed control for the traffic lights, and with traffic lights controlled according to the max pressure rule. These simulations show that the proposed local MPC controller achieves a significant reduction in delay for various traffic conditions. [ABSTRACT FROM AUTHOR]

Highlights • Propose a multimodal dynamic user equilibrium model with MFD traffic dynamics. • Derive the optimal bus fare and service frequency for given space allocation in steady state. • Discover the counter-intuitive properties of traffic equilibrium and optimal bus service. • Examine the interaction between managing authority and service operators under responsive bus service. • Explore the optimal space allocation strategies in dynamic cases with different economic objectives. Abstract A smart design of transport systems involves efficient use and allocation of the limited urban road capacity in the multimodal environment. This paper intends to understand the system-wide effect of dividing the road space to the private and public transport modes and how the public transport service provider responds to the space changes. To this end, the bimodal dynamic user equilibrium is formulated for separated road space. The Macroscopic Fundamental Diagram (MFD) model is employed to depict the dynamics of the automobile traffic for its state-dependent feature, its inclusion of hypercongestion, and its advantage of capturing network topology. The delay of a bus trip depends on the running speed which is in turn affected by bus lane capacity and ridership. Within the proposed bimodal framework, the steady-state equilibrium traffic characteristics and the optimal bus fare and service frequency are analytically derived. The counter-intuitive properties of traffic condition, modal split, and behavior of bus operator in the hypercongestion are identified. To understand the interaction between the transport authority (for system benefit maximization) and the bus operator (for its own benefit maximization), we examine how the bus operator responds to space changes and how the system benefit is influenced with the road space allocation. With responsive bus service, the condition, under which expanding bus lane capacity is beneficial to the system as a whole, has been analytically established. Then the model is applied to the dynamic framework where the space allocation changes with varying demand and demand-responsive bus service. We compare the optimal bus services under different economic objectives, evaluate the system performance of the bimodal network, and explore the dynamic space allocation strategy for the sake of social welfare maximization. [ABSTRACT FROM AUTHOR]

Highlights • Guidelines for evaluating traffic prediction systems are proposed. • A methodology complying with these guidelines is described. • The approach is tested on two traffic prediction systems. Abstract Model-based traffic prediction systems (mbTPS) are a central component of the decision support and ICM (integrated corridor management) systems currently used in several large urban traffic management centers. These models are intended to generate real-time predictions of the system's response to candidate operational interventions. They must therefore be kept calibrated and trustworthy. The methodologies currently available for tracking the validity of a mbTPS have been adapted from approaches originally designed for off-line operational planning models. These approaches are insensitive to the complexity of the network and to the amount and quality of the data available. They also require significant human intervention and are therefore not suitable for real-time monitoring. This paper outlines a set of criteria for designing tests that are appropriate for the mbTPS task. It also proposes a test that meets the criteria. The test compares the predictions of the mbTPS in question to those of a model-less alternative. A t-test is used to determine whether the predictions of the mbTPS are superior to those of the model-less predictor. The approach is applied to two different systems using data from the I-210 freeway in Southern California. [ABSTRACT FROM AUTHOR]

Highlights • Novel visualization methods applied to multi-objective signal timing optimization problems. • Metrics developed to quantify level of tradeoff between objectives. • Methods proposed to identify and remove redundant objectives. • Dimensionless variables proposed to combine objectives for final solution ranking. Abstract Adjusting traffic signal timings is a practical way for agencies to manage urban traffic without the need for significant infrastructure investments. Signal timings are generally selected to minimize the total control delay vehicles experience at an intersection, particularly when the intersection is isolated or undersaturated. However, in practice, there are many other potential objectives that might be considered in signal timing design, including: total passenger delay, pedestrian delays, delay inequity among competing movements, total number of stopping maneuvers, among others. These objectives do not tend to share the same relationships with signal timing plans and some of these objectives may be in direct conflict. The research proposes the use of a new multi-objective optimization (MOO) visualization technique—the mosaic plot—to easily quantify and identify significant tradeoffs between competing objectives using the set of Pareto optimal solutions that are normally provided by MOO algorithms. Using this tool, methods are also proposed to identify and remove potentially redundant or unnecessary objectives that do not have any significant tradeoffs with others in an effort to reduce problem dimensionality. Since MOO procedures will still be needed if more than one objective remains and MOO algorithms generally provide a set of candidate solutions instead of a single final solution, two methods are proposed to rank the set of Pareto optimal solutions based on how well they balance between the competing objectives to provide a final recommendation. These methods rely on converting the objectives to dimensionless values based on the optimal value for each specific objectives, which allows for direct comparison between and weighting of each. The proposed methods are demonstrated using a simple numerical example of an undersaturated intersection where all objectives can be analytically obtained. However, they can be readily applied to other signal timing problems where objectives can be obtained using simulation outputs to help identify the signal timing plan that provides the most reasonable tradeoff between competing objectives. [ABSTRACT FROM AUTHOR]

Highlights • Demonstrates MFD uncertainty changes network dynamics. • Markov Chain framework developed that addresses uncertainty in MFD and vehicle arrivals. • Analytical Markov Chain model accurately estimates dynamics under uncertainty. • Markov chain model applicable to single- or multi-region networks. Abstract Recent studies have proposed using well-defined relationships between network productivity and accumulation—otherwise known as Network or Macroscopic Fundamental Diagrams (MFDs)—to model the dynamics of large-scale urban traffic networks. This provides a computationally efficient way to study these complex systems and facilitates the design and control of novel large-scale traffic management strategies. However, empirical and simulation evidence suggests that MFDs are rarely well-defined. Instead, they exhibit large amounts of scatter and uncertainty, which suggests a range of network productivities may be observed for any given accumulation. This paper examines the impact of this MFD uncertainty and uncertainty in aggregate-level vehicle demands (i.e., vehicle exit and entry rates) on large-scale network behavior. It is shown that these uncertainties can cause fundamentally different aggregate network behaviors than would be expected if they were ignored, including unexpected congestion or gridlock. An analytically derived Markov Chain framework is proposed that can be used to model aggregate network dynamics while explicitly accounting for these types of uncertainties, which are very likely to arise on realistic urban networks. Comparison between the analytical predictions and numerical simulations suggest that the Markov Chain framework can accurately predict traffic dynamics under uncertainty for both single- and multi-region networks. Since this framework relies on the careful discretization of both time and accumulation within individual regions of a network, guidance is also provided on how to best select these discretization parameters the most accurate results. [ABSTRACT FROM AUTHOR]

Highlights • Network features are leveraged to study incident-induced impacts on road networks mobility. • Four network features are extracted to reflect the distinctive functionality of urban intersections. • Incidents impacts are measured in both temporal and spatial dimension. • Temporally, accident delay is significantly correlated with the Betweenness Centrality and K-shell. • Spatially, micro impact and macro impact are found to be strongly associated with the four network features. Abstract In this paper, we seek to investigate the spatiotemporal impacts of traffic incident on urban road networks. The theoretical lens of a complex network leads us to expect that incident impacts are associated with the functionality that an intersection acts in a network, and also, the location of incident sites. Incident impacts are measured in both temporal and spatial dimension through mining the large-scale traffic flow data in conjunction with the incident record. In the complex network context, the urban road network can be converted into a weighted direct graph with intersections as nodes and road segments as edges with their geographic information. Four network features, i.e., Betweenness Centrality, weighted PageRank, Hub, and K-shell are assigned to each intersection to measure its functionality. Temporally, we find out significant correlations between incident delay and two network features by applying hazard-based models. Spatially, the micro impact and the macro impact are found to be strongly associated with three network features through estimating a Bayesian Negative-binomial Conditional Autoregressive model and a generalized linear model, respectively. Our study provides the basis of leveraging urban road network context to evaluate incident impacts, with some explanations, insights and possible extensions that would assist traffic administrations to guide the post-incident resilience and emergency management. [ABSTRACT FROM AUTHOR]

Road network congestion; a traffic state characterised by slower speeds, longer trip times, and increased vehicular queuing; is a major issue in most urban areas around the globe. Building more roads is a commonly employed policy intervention to reduce congestion. This strategy, however, is controversial because under certain conditions road capacity expansions may induce growth in traffic volumes. A crucial precursor to understanding whether road capacity expansions provide a solution to congestion is to quantify the technology driving congestion in urban road networks. This congestion technology describes the variation in performance of the network, often represented by traffic flow through the road network, over its intensity of use given by the number of vehicles in the network. However, obtaining empirical estimates of congestion technology from data on traffic variables is challenging due to statistical biases that emerge via the complex interactions between traffic flow, traffic controls, and capacity. To adjust for such biases, this paper presents an approach based on causal statistical modelling to quantify the nature and form of congestion technology in road networks in twenty-four cities worldwide. Our results suggest that increasing network capacity is in general not an efficient solution to manage congestion, in the sense that the average travel speed in the network does not increase substantially with an increase in capacity. This result and our congestion technology estimates have important implications for optimal urban transportation strategies. [ABSTRACT FROM AUTHOR]

With rapid population growth and urban development, traffic congestion has become an inescapable issue, especially in large cities. Many congestion reduction strategies have been proposed in the past, ranging from roadway extension to transportation demand management. In particular, congestion pricing schemes have been used as negative reinforcements for traffic control. In this paper, we study an alternative approach of offering positive incentives to drivers to take different routes. More specifically, we propose an algorithm to reduce traffic congestion and improve routing efficiency via offering (personalized) incentives to drivers. We propose to exploit the wide accessibility of smart devices to communicate with drivers and develop an incentive offering mechanism using individuals' preferences and aggregate traffic information. The incentives are offered after solving a large-scale optimization problem in order to minimize the total travel time (or minimize any cost function of the network such as total carbon emission). Since this massive-size optimization problem needs to be solved continually in the network, we developed a distributed computational approach. The proposed distributed algorithm is guaranteed to converge under a mild set of assumptions that are verified with real data. We evaluated the performance of our algorithm using traffic data from the Los Angeles area. Our experiments show congestion reduction of up to 5% in arterial roads and highways. • Reducing congestion by improving routing efficiency. • Offering incentives to reduce congestion using individuals' preferences of drivers. • A distributed computational approach to incentive offering. • The methods are evaluated on a large-scale real-world dataset. [ABSTRACT FROM AUTHOR]

• A novel multi-agent gated collaborative signal control method is proposed. • Combining Bi-LSTM and attention establishes communication between agents. • The reward is developed to trade off the macro and micro control contradictions. • Simulation experiments verify the effectiveness of the proposed control model. 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. [ABSTRACT FROM AUTHOR]

Recent technological advancements provide a level of mobility never seen before to modern societies. Sustaining today’s economy and societies depend on maintaining this mobility. However, this mobility also causes undesirable effects, especially in high urban population areas. Even though, higher priority is given to public transport in metropolitan areas, road network is still an important part of everyday commute and should be planned and managed with great care. In this study, we propose an optimisation methodology that would help the traffic authorities to better predict the results of strategic management decisions in a realistic traffic model. We formulate a bi-level multi-objective traffic optimisation model with a sustainability perspective. The upper level of the proposed model considers the traffic authority’s management strategies while the lower level considers the traffic users’ decisions. The lower level is modelled using the Stochastic User Equilibrium since it allows more realistic results than the deterministic one. A case study is provided to illustrate the proposed model. The proposed methodology provides an avenue for understanding the trade-offs among conflicting objectives and for designing an environmentally and socially sustainable transportation system. More importantly, it builds the foundation for an intelligent traffic management decision support system. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we study the application of perimeter flow control by simulation of cellular automaton model on urban traffic system. We find that the relation of traffic flow and vehicle density (Macroscopic Fundamental Diagram, MFD) will have different shapes for the core area and the peripheral area. The MFD shows free-flow state, saturate flow state and congestion state. But the magnitude of maximal flow and the position for congestion transition are different for the whole system and the core area. We suggest to realize the perimeter control strategy by assigning a special prohibiting phase to the perimeter traffic lights for the roads entering the core area. The strategy is controlled by two critical densities ρ 1 and ρ 2 , which are determined by the MFD of the core area. Simulations show that both the average arrival rate and the average flow will be greatly improved with the perimeter flow control strategy. In addition, the perimeter flow control strategy can increase the critical density of traffic congestion. The probability of system-wide gridlock will decrease, and the system can perform well under both close and open boundary conditions. All the results indicate that the perimeter flow control strategy can effectively improve the performance of the traffic system. Highlights • MFD has different shape for the system and for the core area. • Parameters of perimeter control should be determined by the MFD of core area. • Perimeter control can reduce gridlock and increase the critical density. [ABSTRACT FROM AUTHOR]

Highway traffic control signs are commonly used to regulate, warn, and guide road users. It is widely believed that traffic signs comprehension has a tremendous effect on traffic safety. The primary objective of this research is to investigate the relationship between drivers’ personal characteristics and their familiarity/comprehensibility with a thirty nine posted traffic signs. To this end, 400 surveys were distributed among Jordanian drivers. The results showed that the familiarity level of traffic signs is higher than comprehensibility level. On average 79%, 77%, and 83% of the drivers were familiar with regulatory, warning, and guidance traffic signs, respectively. On the other hand, only 61%, 66%, and 75% of the drivers comprehended regulatory, warning, and guidance traffic signs, respectively. “Narrow Bridge”, “Divided Roadway a Head”, “Dead End” and “Highway” received the lowest comprehensibility scores among drivers. Participants with commercial driving license had higher familiarity and comprehensibility levels than those with regular license. Drivers with a driving experience more than 11years show more familiarity and comprehensibility for traffic signs than those with less than 2 years driving experience. The number of traffic violations did not have a significant effect on traffic signs familiarity and comprehensibility. [ABSTRACT FROM AUTHOR]

This paper reports a review of the extensions and application of the Cell Transmission Model (CTM). Those extensions are models able to simulate complex urban traffic dynamics with all the advantages of macroscopic and microscopic urban traffic model. Over the past few years researchers have been trying to increase the level of detail by extending CTM and introducing new formulations to improve the application of the model in urban traffic. The authors classified the papers while taking into consideration all those factors characterizing the urban traffic, arterial and intersection traffic flow in particular. One of the primary goals of transport research is to develop a general framework of urban traffic networks that might be applied from a realistic point of view. Recent studies about traffic simulations have shown that, among various macroscopic simulation models, the CTM has the potential to achieve this objective. We have also reported our model the CTM_UT that improves the CTM for Urban Traffic. We believe that it is possible to apply this model to ITS application, hence increase the accuracy of the macroscopic model while maintaining the computational advantages and provide an accurate prediction of travel time approach. [ABSTRACT FROM AUTHOR]

Perimeter control based on the Macroscopic Fundamental Diagram (MFD) is widely developed for alleviating or postponing congestion in a protected region. Recent studies reveal that traffic conditions might not be improved if the perimeter control strategies are applied to unstable systems where high demand generates heavy and heterogeneously distributed traffic congestion. Therefore, considering stability of the targeted traffic system is essential, for the sake of developing a feasible and then optimal control strategy. This paper sheds light on this direction. It integrates a stability characterization algorithm of MFD system equations into the Model Predictive Control (MPC) scheme, and features respectively an upper and a lower bound of the feasible control inputs, to guarantee system stability. Firstly, the dynamics of traffic heterogeneity and its effect on the MFD are analyzed, using real data from Guangzhou in China. Piecewise affine functions of average flow are proposed to capture traffic heterogeneity in both regional and subregional MFDs. Secondly, stability of a three-state two-region system is investigated via stable equilibrium and surface boundaries analysis. Finally, a three-layer hierarchical control strategy is introduced for the studied two-region heterogeneous urban networks. The first layer of the controller calculates the stable surface boundaries for the given traffic demands and then determines the bounds of control input (split rate). An MPC approach in the second layer is used to solve an optimization problem with two objectives of minimizing total network delay and maximizing network throughput. Heterogeneity among the subregions is minimized in the last layer by implementing simultaneously a subregional perimeter flow control and an internal flow control. The effectiveness and stability of the proposed control approach are verified by comparison with four existing perimeter control strategies. [ABSTRACT FROM AUTHOR]

In this paper, a model-based perimeter control policy for large-scale urban vehicular networks is proposed. Assuming a homogeneously loaded vehicle network and the existence of a well-posed Network Fundamental Diagram (NFD), we describe a protected network throughout its aggregated dynamics including nonlinear exit flow characteristics. Within this framework of constrained optimal boundary flow gating, two main performance metrics are considered: (a) first, connected to the NFD, the concept of average network travel time and delay as a performance metric is defined; (b) second, at boundaries, we take into account additional external network queue dynamics governed by uncontrolled inflow demands. External queue capacities in terms of finite-link lengths are used as the second performance metric. Hence, the corresponding performance requirement is an upper bound of external queues. While external queues represent vehicles waiting to enter the protected network, internal queue describes the protected network’s aggregated behavior. By controlling the number of vehicles joining the internal queue from the external ones, herewith a network traffic flow maximization solution subject to the internal and external dynamics and their performance constraints is developed. The originally non-convex optimization problem is transformed to a numerically efficiently convex one by relaxing the performance constraints into time-dependent state boundaries. The control solution can be interpreted as a mechanism which transforms the unknown arrival process governing the number of vehicles entering the network to a regulated process, such that prescribed performance requirements on travel time in the network and upper bound on the external queue are satisfied. Comparative numerical simulation studies on a microscopic traffic simulator are carried out to show the benefits of the proposed method. [ABSTRACT FROM AUTHOR]

Nowadays, traffic management has become a challenge for urban areas, which are covering larger geographic spaces and facing the generation of different kinds of traffic data. This article presents a robust traffic estimation framework for highways modeled by a system of Lighthill Whitham Richards equations that is able to assimilate different sensor data available. We first present an equivalent formulation of the problem using a Hamilton–Jacobi equation. Then, using a semi-analytic formula, we show that the model constraints resulting from the Hamilton–Jacobi equation are linear ones. We then pose the problem of estimating the traffic density given incomplete and inaccurate traffic data as a Mixed Integer Program. We then extend the density estimation framework to highway networks with any available data constraint and modeling junctions. Finally, we present a travel estimation application for a small network using real traffic measurements obtained obtained during Mobile Century traffic experiment, and comparing the results with ground truth data. [ABSTRACT FROM AUTHOR]