Search

The multi-agent environment is now widely recognised as a key design abstraction for constructing multi-agent systems, equally important as the agents. An explicitly designed environment may have several roles, such as the inter-mediation between agents, the support for interaction, the embodiment of rules and constraints, etc. Mobility applications fit perfectly with a design in the form of a multi-agent system with an explicit environment model. Indeed, in these applications, the components of the system are autonomous and intelligent (drivers, travellers, vehicles, etc.), and the transportation network is a natural environment that they perceive and on which they act. However, the concept of the multi-agent environment may be profitably used beyond this specific geographical context. This paper discusses the relevance of the multi-agent environment in mobility applications and describes different use cases in simulation and optimisation.

This study proposes a new framework to solve the road network evacuation problem, considering a dynamic allocation of evacuees to shelters. Although many studies have been performed on this problem with static settings, there are few studies in the literature that address this problem in a dynamic context. We couple and solve the dynamic traffic assignment (DTA) and dynamic shelter allocation problem (SAP) using agent-based dynamic simulation. The model for the SAP aims to satisfy system operator interests by determining the best shelters for evacuees, and evacuees tend to reach their shelters as fast as possible. We validate our methodology on the real network of Luxembourg City and evaluate its performance in front of one of the advanced methods in the literature. The results show that calculating dynamic shelter allocation improves mean evacuation time and significantly reduces the network clearance time compared to the methods with fixed shelter allocation plans.

One of the most challenging facets of dynamic ride-sharing problems is the automated matching process, especially in large-scale networks where the number of received requests is huge. Matching users to trips is very challenging in real-time since it must happen very quickly while it must ensure high quality solutions. In this paper, we rely on an exact solving method to ensure the quality of the solution, while using AI-based techniques to limit the number of requests that we feed to the solver. More precisely, we propose a clustering method based on a new space-time shareability function to put the most shareable trips inside separate clusters. Each cluster is then managed with a proposed heuristic framework in order to solve the matching problem inside each cluster. To validate our proposal, we employ the proposed method on the network of Lyon city in France, with half-million requests in the morning peak from 6 to 10 AM. The results demonstrate that the algorithm can provide high-quality solutions in a short time for large-scale problems.

In their daily activity planning, travelers always considers time and space constraints such as working or education hours and distances to facilities that can restrict the location and time-of-day choices of other activities. In the field of population synthesis, current demand models lack dynamic consistency and often fail to capture the angle of activity choices at different times of the day. This article presents a method for synthetic population generation with a focus on activity-time choice. Activity-time choice consists mainly in the activity’s starting time and its duration, and we consider daily planning with some mandatory home-based activity: the chain of other subsequent activities a traveler can participate in depends on their possible end-time and duration as well as the travel distance from one another and opening hours of commodities. We are interested in a suburban area with sparse data available on population, where a discrete choice model based on utilities cannot be implemented due to the lack of microeconomic data. Our method applies activity-hours distributions extracted from the public census, with a limited corpus, to draw the time of a potential next activity based on the end-time of the previous one, predicted travel times, and the successor activities the agent wants to participate in during the day. We show that our method is able to construct plannings for 126k agents over five municipalities, with chains of activity made of work, education, shopping, leisure, restaurant and kindergarten, which fit adequately real-world time distributions.

KES-AMSTA 2021, 15th KES International Conference Agents and Multi-Agent Systems: Technologies and Applications, Virtuel, , 14-/06/2021 - 16/06/2021; Mobility simulations are an effective tool to forecast the effects of transportation policies. They are a useful part of decision support systems for policymakers. Big real-estate projects, aiming at creating whole new neighborhoods, for instance, need this kind of application to estimate their impacts on the surrounding environment. The potential visits of the various equipments and the public spaces of the new neighborhood can also be estimated with this kind of tools. In this context, agent-based simulation is a relevant model for the design and implementation of transportation applications, that represents travelers individually, together with their perception model of the environment and their internal decision processes. In this paper, we propose a multi-agent activity-based simulation to represent current and future mobility patterns in a city, with or without a new neighborhood. We describe the inputs of the simulation, in terms of networks and travel demand, the behaviors of the agents, and the outputs on the laVallée project (France). The first series of experiments demonstrate the potential of the simulation and its benefits for the project managers and the decision-makers in the concerned territories.

Can dynamic ride-sharing reduce traffic congestion? In this paper we show that the answer is yes if the trip density is high, which is usually the case in large-scale networks but not in medium-scale networks where opportunities for sharing in time and space become rather limited. When the demand density is high, the dynamic ride-sharing system can significantly improve traffic conditions, especially during peak hours. Sharing can compensate extra travel distances related to operating a mobility service. The situation is entirely different in small and medium-scale cities when trip shareability is small, even if the ride-sharing system is fully optimized based on the perfect demand prediction in the near future. The reason is simple, mobility services significantly increase the total travel distance, and sharing is simply a means of combating this trend without eliminating it when the trip density is not high enough. This paper proposes a complete framework to represent the functioning of the ride-sharing system and multiple steps to tackle the curse of dimensionality when solving the problem. We address the problem for two city scales in order to compare different trip densities. A city scale of 25 km 2 with a total market of 11,235 shareable trips for the medium-scale network and a city scale of 80 km2 with 205,308 demand for service vehicles for the large-scale network over a 4-hour period with a rolling horizon of 20 minutes. The solutions are assessed using a dynamic trip-based macroscopic simulation to account for the congestion effect and dynamic travel times that may influence the optimal solution obtained with predicted travel times. This outperforms most previous studies on optimal fleet management that usually consider constant and fully deterministic travel time functions.

Identifying the spatio-temporal patterns of people activities in urban areas is key to effective urban planning, it can be used in real-estate projects to predict their future impacts on behavior in surrounding accessible areas. LaVall&eacute, e is a large construction project recently started in Paris&rsquo, s suburb, it is a new district due in 2024. The paper is in the field of urban planning, aiming at developing a method making it possible to model the potential visits of the various equipment and public spaces of the district, by mobilizing data from census at the departmental level, and the layout of shops and activities as defined by the real-estate project. This model takes into account the flow of external visitors, estimated realistically based on the pre-project movements in the areas of influence of LaVall&eacute, e. In this paper, we propose an activity-based model methodology to determine trips and their purpose at a mesoscopic scale including the city and surrounding areas, in the current baseline scenario. This travel demand is required to estimate potential external visitors of the future district. A first demonstration shows that the model correctly represents the current demands and allows the forecast of future demand in the area.

hEART 2019, 8th Symposium of the European Association for Research in Transportation, Budapest, HONGRIE, 04-/09/2019 - 06/09/2019; In recent years, intelligent transportation systems made it possible for operators to adapt in real-time the transportation supply to travel demand via new mobility services. Among these services, ride-sharing is becoming popular. The dynamic ride-sharing problem involves two sub-problems: (1) How to serve the upcoming trips (Optimal fleet management) and (2) How to accurately predict the travel times to determine vehicles availability and pick up/drop off times. In this paper, we express the optimal fleet management problem as a constrained multi objective integer linear programming. Our aim is to find the global optimal solution for the ride-sharing problem without uncertainty in demand to have a vision of these services performance in optimal situation. Then we compare the results and experiments with the exact optimal condition. We have designed an algorithm to find the exact solution for matching problem based on the branch and bound algorithm. To solve the second sub-problem, we define two different models to assess the impact of traffic conditions on the dynamic ride-sharing system performance for large-scale problems.

Mes travaux de recherche portent sur les systèmes multi-agents et leur application aux problèmes et applications de transport dynamiques. Je suis intéressé par le développement de modèles, de simulations et d'algorithmes pour aider les acteurs du transport multimodal à résoudre les problèmes complexes induits par les changements continus dans l'offre et la demande de services de transport. Je propose un certain nombre de modèles et de simulations aidant les acteurs du transport à observer, estimer et optimiser l'état des réseaux de transport. Mon travail contribue à une meilleure compréhension des problèmes soulevés par les nouveaux services de mobilité (numérotation, covoiturage, systèmes de recherche de parkings urbains, etc.) et à une meilleure conception des systèmes d'exploitation et des systèmes d'information du voyageur, intégrant ces nouveaux services., My research work deals with multi-agent systems and their application to dynamic transportation problems and applications. I am interested in the development of models, simulations and algorithms to support the multimodal transportation actors solving the complex problems induced by the continuous changes in the supply and demand of transportation services. I propose a number of models and simulations supporting the transportation actors in observing, estimating and optimizing the state of the transportation networks. My work contributes to a better understanding of the problems raised by the new mobility services (dial a ride, ride sharing, urban parking search systems, etc.) and to a better design of operating systems and traveler information systems, integrating these new services.

TRISTAN X, The Tenth Triennial Symposium on Transportation Analysis, Hamilton island, AUSTRALIE, 17-/06/2019 - 21/06/2019; Ride-sharing is becoming one of the most important transportation modes in many cities due to its convenience and accessibility. But it is important to assess the impact of network on the ride-sharing performance and vice versa. This paper proposes a complete framework to represent the ride-sharing system functioning. We address the problem for a city scale of 25 km2, a total market of 11,235 shareable trips over a 4 hours period with a rolling horizon of 20 minutes. Furthermore, the solutions are assessed using a dynamic trip-based macroscopic simulation to account for congestion effect and dynamic travel times that may influence the optimal solution obtained with predicted travel times. This outperforms most previous studies about optimal fleet management that usually consider constant and fully deterministic travel time functions. Then we assess the network performance under different levels of ride-sharing.

Les autres actes de conférences de l'ESAW sont publiées par Springer dans la série LNAI: "Engineering Societies in the Agents World IX"; In this paper, we propose a multi-agent approach to solve a hard combinatorial problem: the vehicle routing problem with time windows. A multi-agent approach is most relevant when dealing with the dynamic version of the problem, where all the customers are not known before the start of the solving process. We focus on both the interaction process and the computation process. For the interaction process, we rely on the Property-Based Coordination model, enabling to limit the interaction to the only agents that might reach a solution. The overall interaction follows the well-known Contract Net Protocol. Concerning the computation process, we propose a new heuristic, based on the measure of the agents’ perception field over their environment, which privileges a solving process focused on the future.