Multi-Agent Activity-Based Simulation of a Future Neighborhood - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Multi-Agent Activity-Based Simulation of a Future Neighborhood

Résumé

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.

Mots clés

Fichier principal
Vignette du fichier
tex00004990.pdf (993.45 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03363823 , version 1 (04-10-2021)

Identifiants

  • HAL Id : hal-03363823 , version 1

Citer

Younes Delhoum, Rachid Belaroussi, Francis Dupin, Mahdi Zargayouna. Multi-Agent Activity-Based Simulation of a Future Neighborhood. KES-AMSTA 2021, 15th KES International Conference Agents and Multi-Agent Systems: Technologies and Applications, Jun 2021, Virtuel, France. pp. 501-510. ⟨hal-03363823⟩
24 Consultations
131 Téléchargements

Partager

Gmail Facebook X LinkedIn More