Simon Pageaud, Véronique Deslandres, Salima Hassas, Vassilissa Lehoux
JFSMA, Métabief, France, 10 - 12 October 2018
In the near future, the increasing availability of data will require policy makers to regularly change urban policies to incorporate changing behavior and user feedbacks. In this paper, we propose a generic agent-based architecture for designing and modeling urban policies in order to test their relevance by deploying them on a specific environment. Environments are designed using data from the city provided by any available open data toolkit (Open Street Map here). Two agent-based models are coupled in a micro-macro dynamic loop and they can be adapted either by the system using reinforcement learning or by the stakeholders using simulation results. We propose a formalism to represent urban policies to allow a co-designed iterative process between the policymaker and our system. An experimentation on the regulation of parking prices in downtown area justifies the use of our architecture to design relevant urban policies.​
Report number: