Skip to main content

A Multiscale and Multimodal Modelling Approach for Green Urban Traffic Management

Periodic Reporting for period 4 - MAGnUM (A Multiscale and Multimodal Modelling Approach for Green Urban Traffic Management)

Reporting period: 2020-03-01 to 2021-02-28

The MAGnUM project aims to (i) create a consistent set of interrelated dynamic and multimodal traffic models able to capture driver behaviors at the different urban scales and (ii) apply this variety of models to design efficient and green traffic management strategies.
Traffic flow dynamics are well reproduced at a local urban scale by the kinematic wave model and its numerous extensions. Even if this model is parsimonious compared to other modeling approaches, it can hardly be applied at larger urban scales for traffic control applications. It is challenging to design transportation models that address the full urban scale, account for all multimodal options and new mobility services while considering user choices and adaptations to new options.
Nevertheless, it is critical to envision significant breakthroughs in improving urban mobility management while decreasing its environmental footprint. It is the core ambition of the MAGnUM project. From the modeling point of view, the core idea is to better focus on user trips than network flow when designing aggregate dynamic modeling. This primary goal is achieved by mixing analytical investigations on idealized but insightful test cases with explanatory approaches using data gathered from dynamic simulations, serious game sessions, and actual observations on more realistic and complex cases.
The second goal of the project concerns the design of innovative traffic management strategies at multiple urban scales. Breakthroughs have been achieved by considering multiple and competitive objectives when optimizing with a tight focus on environment issues and multi-modality.
The MAGnUM project has resulted in several breakthroughs in both urban transportation system modeling and control. A better characterization of user choices was achieved through an innovative data collection method: simulation game sessions. We designed a new generation of dynamic multimodal traffic models that provide fast prediction at a large-scale while keeping track of user heterogeneities and ensuring consistency with local traffic dynamics. Extensive numerical and experimental studies lead to cutting-edge methods for model calibration from conventional and emerging data sources and showed promising validation results. New computational methods were proposed to speed up network equilibrium calculations and address large urban scale problems considering both route, mode, and departure time choices. We designed a new generic framework to unveil the most penalizing trips considering the global network performance. We developed a new traffic management framework based on optimal route guidance that guarantees user privacy by design. The idea is to broadcast avoidance maps to vehicle navigation systems that can then self-adapt to improve urban traffic conditions collectively. This concept will be tested in fall 21 in Lyon Metropolis in a follow-up ERC PoC project. Lastly, we improved existing traffic management strategies and designed new green solutions. These include improving bus operations, different solutions to move the network towards system optimum through incentive or penalties, advances in perimeter control considering user adaptations through route choices, and investigations on ride-sharing benefits. One salient result is a new framework to twist the original perimeter control concept to minimize total emission at the city scale instead of only improving traffic conditions inside the perimeter.
The main methodological contributions of the MAGnUM project are related to three interrelated topics.

Network traffic modeling: Earlier developments for large-scale network models partition a city in multiple regions and track flow exchanges to represent traffic dynamics. We proposed a new methodological framework that keeps the parsimony of regional traffic dynamics but individualizes trips inside the multimodal transportation system. This trip-based approach is proven more accurate in free-flow and saturated traffic. To adequately handle congestion spillbacks in oversaturated regimes, we mix trip-based with the original accumulation-based approach at the region perimeters. Our framework can handle multiple trip lengths in each region. We also developed new methods for calibrating large-scale models considering innovative datasets like mobile phones or drone footages.

Network equilibrium and traffic assignment: We introduced a new regional framework to handle route choices across regional paths. The main challenge was scaling-up the road network characteristics into relevant aggregate figures. In particular, we design methods to identify the prevailing regional paths and estimate their lengths considering both the network topology and the local routing discipline due to congestion. We also revisited the network equilibrium problem by introducing metaheuristics (simulating annealing and genetic algorithms) from the optimization field. Such a computational framework permits parallelizing some calculations and leads to significant time savings while improving accuracy. We then propose a new framework to identify trips that reduce the most network performance. Based on machine learning technics, we characterize these trips only based on their features and network characteristics. It paves the way for their direct identification in real-time. Lastly, a significant methodological contribution is the introduction of Mean Field Game (MFG) theory to solve the optimal departure time problem. Classical methods suffer from scaling issues that prevent them from addressing large-scale samples. They also resort to restrictive assumptions that reduce the solution space exploration. Combining MFG with the trip-based modeling framework blew up those restrictions and efficiently solved realistic test cases with more than 40000 trips.

User behaviors, data collection, and analysis: The simulation game we developed provides priceless behavioral observations. It cannot replace on-field experiments, but it simultaneously collects information for many users considering multiple network configurations. Furthermore, as the simulation is fully controlled and finely tuned, it permits to have a perfect vision of the local and global network scales. Based on such unprecedented datasets, we performed thorough data analysis and implemented advanced clustering technics to define the most representative OD pairs and investigate bounded rationality at the individual level.
A cutting-edge contribution in data analysis is the concept of 3D congestion maps. We benchmarked different clustering algorithms to reduce the link speeds variation into a small bunch of homogeneous space-time domains. We were also able to cluster days with homogeneous patterns and apply consensual learning to determine each cluster's typical congestion map. Our test case, i.e. Amsterdam's city, shows that only four consensual maps can describe the city traffic dynamics over 35 days. We also studied two other datasets in Zurich and Luzerne over an entire year. Here, we use a new similarity measure for MFD based on dynamic time warping to characterize and compare traffic states' daily time evolution. Again, a few consensual patterns – 7 - suffice to describe all days. We related these patterns to different bottleneck activation at the city level. It results in an early identification system, which identifies the more likely pattern for the current day.
Simulation game session
Large-scale simulation based on reservoir - screenshot at a given time step
3D congestion maps for the city of Amsterdam (one day)
Large-scale simulation based on reservoir - application to one district of Lyon Metropole