Max-pressure (MP) control has emerged as a prominent real-time network traffic signal control strategy due to its simplicity, decentralized structure, and theoretical guarantees of network queue stability. Meanwhile, advances in connected vehicle (CV) technology have sparked extensive research into CV-based traffic signal control. Despite these developments, few studies have investigated MP control in heterogeneously distributed and partially CV environments while ensuring network queue stability. To address these research gaps, we propose a CV-based MP control (CV-MP) method that leverages real-time CV travel time information to compute the pressure, thereby incorporating both the spatial distribution and temporal delays of vehicles, unlike existing approaches that utilized only spatial distribution or temporal delays. In particular, we establish sufficient conditions for road network queue stability that are compatible with most existing MP control methods. Moreover, we pioneered the proof of network queue stability even if the vehicles are only partially connected and heterogeneously distributed, and gave a necessary condition of CV observation for maintaining the stability. Evaluation results on an Amsterdam corridor show that CV-MP significantly reduces vehicle delays compared to both actuated control and conventional MP control across various CV penetration rates. Moreover, in scenarios with dynamic traffic demand, CV-MP achieves lower spillover peaks even with low and heterogeneous CV penetration rates, further highlighting its effectiveness and robustness.

Smart traffic systems, like those using wellestablished methods such as SCOOT, SCATS and TUC, aim to improve traffic flow by dynamically adjusting signal timings based on real-time traffic conditions. Traffic engineers need to understand the objective functions behind traffic signal control to analyze, improve, and optimize network performances. However, different jurisdictions, different operators and competing interests imply that the underlying objective functions governing traffic signal control might not be publicly known with sufficient detail (e.g. to preserve Intellectual Property Rights). A method for discovering these functions is therefore needed, particularly to enable better cooperation among stakeholders. In this work, we train computer models to mimic the decisions made by smart traffic light systems. Using data from a simulated traffic network (with virtual sensors tracking vehicles), we test a variety of supervised models, ranging from simple decision trees to more complex neural networks. Our results show these models can accurately mimic the underlying system's actions, achieving up to 99% accuracy. This work demonstrates that supervised learning can serve as a powerful tool for uncovering hidden traffic control functions by training models to replicate the system's decisions. By analyzing these models, we can then infer the key factors influencing signal control, thereby gaining insights into the underlying objective function.

Max-pressure (MP) control has proven effective at stabilizing network queues and improving traffic throughput in large-scale urban road networks. However, conventional MP controllers based on connected vehicle (CV) data face two critical limitations: network stability diminishes when connected vehicle (CV) penetration rates are low, and significant privacy concerns arise when utilizing individual vehicle data. To address these challenges, this paper proposes a novel Private-MP controller that fuses data from both fixed-location detectors and CVs in an architecture of mobile edge computing. To fully safeguard CV privacy, including macro-route information and micro-trajectory information, Private-MP employs a privacy-preserving mechanism that combines homomorphic encryption with an adaptive randomized response strategy. Simulation studies on a network with five intersections showed that despite some increases in average vehicle delay due to privacy protection, Private-MP still ensures a more robust performance on average vehicle delay than CV-based MP in low penetration rate scenarios and outperforms traditional detector-based MP control while improving fairness among connected and non-connected vehicles.

Large Language Models zijn AI-systemen die menselijke taal begrijpen en zich er ook in kunnen uiten. Ze zijn de basis onder populaire applicaties als ChatGPT, Gemini en Copilot. Maar inmiddels is de technologie zó breed inzetbaar dat ze ook doordringt in de mobiliteitssector. Hoe werken de Large Language Models? Hoe kunnen ze van nut zijn in ons vakgebied? En wat zijn de mitsen en maren

This study investigates the potential of link-and path-based incentives to mitigate congestion in urban transportation networks. Both incentive schemes are formulated as non-linear optimisation problems with complementarity constraints. Mathematically, it is demonstrated that the feasible region of the link-based model is a subset of the feasible region of the path-based model. Consequently, path-based incentives exhibit greater potential for shifting the user equilibrium flow pattern toward system optimum compared to link incentives. A column generation-based iterative solution technique, which generates new paths at each iteration, is devised to efficiently solve both optimisation problems. Numerical experiments conducted for various transport networks also highlight the superiority of path-based incentives in reducing total travel time in urban transportation networks.

Among real-time traffic control methods, max-pressure (MP) control stands out due to its simplicity, decentralized nature, and robust theoretical foundation. Besides, advancements in connected vehicle (CV) technology have motivated a significant amount of research into traffic signal control based on CVs. Nevertheless, few studies have been dedicated to MP control in partially CV environments and meanwhile consider multi-modal traffic flows. To fill this research gap, this study proposes CV-based multi-modal MP control (CV-MMP), which calculates the pressure based on travel time information of CVs weighted by vehicle occupancy. Therefore, a hierarchical multi-modal traffic signal priority controller is achieved in a soft manner. Besides, adapting to the requirements of practical applications, CV-MMP is extended to fuse detector data and consider phase switching lost time and cyclic phase sequence. The evaluation results based on a toy network simulation demonstrate that CV-MMP can significantly reduce transit delay with a small increase in private vehicle delay, resulting in a significant reduction in average person delay. In addition, approximately 75% of CBs pass through the network without experiencing delays due to stopping. Therefore, our method can achieve effective transit signal priority and even transit signal coordination under single transit requests.

This paper presents a novel auction-based traffic signal control mechanism aimed at optimizing multimodal traffic flow at signalized intersections through \textcolor{red}{connected vehicles}. The proposed framework, which utilizes a second price sealed bid auction mechanism, allocates green time dynamically based on user bids, incorporating policy-oriented modal priority. This approach addresses the limitations of current signal control systems by providing a computationally fast and distributable method that considers the priority hierarchy of traffic modes, thereby enhancing the efficiency and equity of intersection management.

Key innovations include a dynamic bid distance determination method and a modified bidding scheme that prioritize certain traffic modes according to predefined policies. The effectiveness of these methods is demonstrated through a case study focusing on bicycle prioritization at a real-world intersection in Bordeaux, France.

Simulation results indicate significant improvements in service levels for prioritized modes without substantially increasing delays for other users. The methods' flexibility for adaptation to different intersection configurations and computational feasibility ensure their applicability to a wide range of intersection types and traffic conditions. Our findings suggest that the sealed bid second price auction mechanism can be a useful tool for policymakers aiming to implement multimodal traffic priorities, contributing to reduced travel delays and more effective control at intersections.

Urban traffic congestion remains a critical challenge in modern cities, with traffic signal control systems often struggling to manage congestion during peak travel times. Perimeter control of a Protected Network (PN) has emerged as a potential solution to reducing gridlock in urban networks. This paper proposes a novel auction-based mechanism for green time allocation at signalized intersections, for effective perimeter control application. Utilising a Sealed Bid, Second Price auction framework, our approach combines real-time traffic monitoring with market-inspired mechanisms to regulate vehicle inflows into PN areas. Unlike existing methods that focus primarily on gated links, our system allocates budgets to individual traffic movements, providing greater flexibility in managing multi-directional flows. We evaluate the proposed mechanism using a test case intersection with a single controlled inflow, comparing it against a volume-based fixed-time approach. The results demonstrate that our auction-based method controls flows into the PN with improved accuracy, outperforming the volume-based approach in terms of inflow regulation, queue management and delays. The framework can be applied in real time to any generic intersection, offering a scalable solution for urban traffic management. This work bridges the gap between perimeter control and market-based intersection auctions, providing a pathway for further research on adaptive traffic management systems.

This study addresses the scalability challenges of the Mixed-Fleet Multi-Terminal Electric Bus Scheduling Problem by exploring various heuristic and metaheuristic approaches applied to large urban networks. A novel Repeated Local Search (RLS) algorithm is developed to optimize full-day scheduling, incorporating key factors such as fleet assignment, charging constraints, and deadheading costs, while accounting for limited charging infrastructure. The RLS method generates initial greedy yet feasible schedules for a mixed fleet of electric and hybrid buses, serving as the foundation for two metaheuristic strategies: Simulated Annealing and a Genetic Algorithm. The Simulated Annealing approach is implemented in two variants: one integrating a Mixed-Integer Linear Programming (MILP)-based move, and the other using an RLS-based move to reschedule trip chains while maintaining feasibility. Meanwhile, the Genetic Algorithm employs repair mechanisms to correct infeasible solutions arising during the crossover process. To evaluate these methodologies, a three-phase experimental framework is employed: (1) stress-testing a MILP model under various fleet and infrastructure conditions, (2) benchmarking MILP performance against metaheuristic methods on small-scale instances, and (3) conducting a comparative analysis of metaheuristics across small, medium, and real-size urban scenarios. The urban-scale instances are derived from real-world public transit timetables in Luxembourg City, encompassing 1,084 trips, 12 terminals, 10 bus lines, and full-day operations. Results indicate that the proposed metaheuristic approaches achieve solutions comparable to exact MILP formulations in small-scale cases while offering substantial scalability improvements for larger networks. Each algorithm exhibits distinct advantages and trade-offs, highlighting the importance of selecting an appropriate method based on the specific scenario and computational constraints. These findings extend prior research on smaller instances and suggest that as urban transit systems transition to electric fleets, the marginal operational benefits for transit agencies may diminish with increasing network size.

Last-mile delivery, one of the most polluting segments of the supply chain, is the focus of numerous studies. There are various innovative delivery methods aimed at mitigating its adverse effects. This study explores whether crowdshipping (CS) could serve as a sustainable urban logistics solution for Rome, assessing its environmental viability. It poses the question: Can Rome adopt CS, and if so, how sustainable would it be? Using real-world data, we employed the MASS-GT simulation tool to simulate parcel demand for various parcel companies in Rome’s urban areas. Additionally, we considered real data on trips made by employees to offices within the study area and their modes of transport. The analyses include predicting parcel demand and forming parcel schedules, both with and without CS.We also assessed employees’ willingness to make detours for parcel pickups. Our findings suggest that CS can reduce emissions depending on users’ willingness to adjust travel routes, which can be incentivized through monetization. Furthermore, by considering the fleet composition of parcel companies, we quantified the potential emissions savings achievable through CS. The results indicate that CS is applicable in Rome and could significantly reduce emissions by approximately 1.3 tonnes of CO2 per day, equivalent to 93 euros in the EU’s Emissions Trading System. This approach aligns with European emissions plans and validates the feasibility of CS in Rome through practical research. It offers valuable insights for policymakers, emphasizing the importance of encouraging user participation and supporting CS platforms.

Multi-modal transport is getting more popular due to the emergence of new traffic
modes. The increase of modes also adds complexity for the transport researchers.
This paper proposes an augmented link-based super-network approach for modeling
multi-modal transport networks, addressing the scalability and versatility issues of
conventional methods. This approach is used to calculate the user equilibrium for
urban transport networks traffic assignment with multiple traffic modes, a difficult
problem due to the intractable enumeration of feasible paths between origindestination
pairs and restricted transfers between different traffic modes. In the supernetwork
representation of multi-modal transport networks, the travel cost of any
feasible route between the origin and destination is formulated as the sum of cost
functions of the augmented links, thus avoiding the enumeration of feasible paths.
Additionally, restrictions on traffic mode transfers can be embedded in the link-based
model by excluding infeasible transfer links or adding penalties for undesired transfers.
The user equilibrium of the augmented link-based super-network model is formulated
as a variational inequality problem, solved using the extra-gradient algorithm. A multimodal
transport network is considered in the case study. Simulation results validate the
effectiveness of the proposed model, demonstrating its scalability and versatility in
addressing complex multi-modal transport networks with diverse traffic modes.
We anticipate that our method can serve as an efficient modeling approach for more
general and complex multi-modal transport networks, facilitating traffic management
and network design.

Are you using tools like ChatGPT in your daily life to help write an email or even draft a construction plan? Just ten years ago, these kinds of capabilities would have seemed unimaginable. Today, they’re becoming part of everyday life for ordinary people. Behind these powerful tools are technologies known as Large Language Models (LLMs)—AI systems that can understand and generate human-like text and now even create images and videos. But what exactly are LLMs? Could they help transform fields like transportation and traffic management? Can they really do everything, or are there still limitations? In this article, we’ll walk you through a general introduction to LLMs: what they are, how they work, and what opportunities—and challenges—they bring to the transportation sector.

Physical distancing has been an important asset in limiting the SARS-CoV-2 virus spread during the COVID-19 pandemic. This study aims to assess compliance with physical distancing and to evaluate the combination of observed and self-reported data used. This research shows that it is difficult to operationalize new rules, that context affects compliance, that there needs to be a need for compliance, and that rules require upkeep. From a methodological point of view, this study found that the combined methods provide a comprehensive picture of compliance behaviour, that it is challenging but essential to mitigate response fatigue in long-term monitoring studies, and that it would be interesting in future research to learn how actual behaviour is influenced by personal narratives.

This study investigates the potential of path-based incentives to mitigate congestion and reduce emissions in urban transport networks using a multi-objective optimisation problem with a budget limit. A column generation-based solution technique is developed that finds a new path between each origin-destination pair at each iteration, and stops when the objective value does not change more than a threshold at two consecutive iterations. Three different scenarios are defined based on the objective function: minimising total travel time (TTT), total emissions (TE), and integrated minimisation of both. Numerical results in the Sioux Falls network show that TTT and TE are conflicting objectives under our modeling assumptions: improving one worsens the other. Nonetheless, the integrated scenario demonstrates the capacity to harmonize both objectives, thereby achieving a reduction in both TTT and TE.

This study investigates the potential of link-and path-based incentives to mitigate congestion and reduce emissions in urban transportation networks. Both incentive schemes are formulated as non-linear optimisation problems with complementarity constraints. Mathematically, it is demonstrated that the feasible region of the link-based model is a subset of the feasible region of the path-based model. Consequently, path-based incentives exhibit a higher potential in pushing the user equilibrium flow pattern toward system optimum, compared to link incentives. A column generation-based iterative solution technique, which generates new paths at each iteration, is devised to efficiently solve both optimisation problems. The numerical results in the Sioux Falls network also highlight the superiority of path-based incentives in reducing total travel time and emissions in urban transportation networks.

The increasing urbanization, combined with shrinking space for transport infrastructure and private parking, significantly challenges urban accessibility. Moreover, the rising number of vehicles exacerbates congestion in city centers, leading to longer commute times, increased noise levels, and greater air pollution. These issues underscore the urgent need for creating low-car urban zones. One promising approach is an integrated traffic management system that considers various modes of transportation—such as cycling, walking, shared mobility, and public transport. However, multi-modal traffic management typically involves diverse stakeholders with potentially conflicting interests, which necessitates a balance of these interests through multi-objective optimization. Traditional approaches often employ a weighted sum method to transform multiple objectives into a single objective. This method significantly constrains the solution space and complicates the assignment of appropriate weights to different objectives. Therefore, generating a Pareto front for multi-modal traffic management could provide decision-makers with a set of efficient solutions, enabling them to select the most suitable option. The ε-constraint method is recognized for its ability to generate a Pareto front. The question we discuss here is whether this method can be effectively applied to managing multi-objective, multi-modal traffic networks. In this study, we answer this question by proposing an augmented ε-constraint-based optimization framework for multi-objective multi-modal traffic management. This framework is bi-level and can accommodate various traffic models and objectives that reflect the diverse interests of multiple stakeholders. Thus the multi-modal traffic management problem can be formulated as a multi-objective nonlinear optimization problem. The augmented ε-constraint method (Mavrotas, 2009) is employed to efficiently address the multiple objectives, and the multi-start sequential quadratic programming method is used to solve the nonlinear optimization problems, such that the Pareto front is obtained. We validate the effectiveness of our framework through a case study, whose preliminary results show that our method improves the traffic performance and provides insights into the trade-off among different objectives.

Our cities are growing at an unprecedented pace. The flexible use of metropolitan infrastructures is the key to maintaining, if not increasing, the current quality of life. The combined use of geo-fence technology and connected vehicles can be the tool to achieve this flexibility. In this paper, we take a first step in the evaluation of the benefits that dynamic geo-fencing could bring. In a simulation-based environment, we employ a computer vision approach to dynamically identify congested areas in a given transportation network. We then compare the performance of perimeter control based on dynamic geo-fencing vs conventional perimeter strategies, based on a fixed, pre-determined area-a scenario mimicking traffic management approaches currently deployed in large metropolitan areas worldwide. Simulation results highlight a reduction of more than 20% of the Total Time Spent in a regular Manhattan grid network, encouraging further efforts to validate the efficiency of dynamic geo-fencing in addressing externalities (congestion, pollution, noise, etc.) in more realistic scenarios.

To use efficiently the infrastructure of transportation networks, control strategies have been developed with the aim to reduce negative externalities, such as congestion and pollutant emissions. Previous works demonstrated that the maximum performance achievable by traffic control policies depends on the number and location of controllers employed, which implies the need to determine a set of controllers capable of fully controlling the underlying transportation network. Various approaches have been explored in the literature to locate controllers on networks, however a gap remains in terms of scalability as the methods proposed often exhibit heavy computational complexity. In this paper we aim to propose an approach capable of locating pricing controllers on transportation networks that is scalable, such that it can be applied on large instances, such as city-sized or regional networks. For this purpose, we propose a topology-based approach, adapted from the sensor location problem, as both problems share similar characteristics. We validate our proposed approach by analyzing the performance of controller sets produced on a wide range of artificially generated network ensembles. The analysis we provide reveals that the method proposed, while being easily applicable on large instances, is capable to locate an efficient controller set, and to redirect flows on the network so as to reduce the total time spent by road users.

We introduce a Markov Modulated Process (MMP) to describe human mobility. We represent the mobility process as a time-varying graph, where a link specifies a connection between two nodes (humans) at any discrete time step. Each state of the Markov chain encodes a certain modification to the original graph. We show that our MMP model successfully captures the main features of a random mobility simulator, in which nodes moves in a square region. We apply our MMP model to human mobility, measured in a library.

In this work we extend our previously proposed cascading Kalman filtering technique, applied to the problem of urban network flow estimation, to adopt heterogeneous traf- fic data sources. Both static infrastructure detection (double induction loops) and Floating Car Data are collected from a given transportation network, and employed within separate stages of the cascading technique. The proposed approach relies upon notions of traffic flow inference (observability) to both i) determine the optimal set of locations in which sensors should be installed and ii) provide enhanced covariance information within the estimation technique. The impact of both penetration rates of Floating Car Data and sensor selection procedure is evaluated empirically, through a microscopic simulation software (SUMO) generating experimental data on a simple grid-like network.
Test results showcase that the proposed extension to the cascading framework is indeed beneficial in reducing the overall estimation error on network segments where static infrastructure is unavailable. Furthermore, the importance of observability- based sensor locations is clearly demonstrated.