Timeline: 2017-2019
Keywords: Demand and Supply Prediction, Probabilistic Graphical Models, Deep Learning.
Team participants: Francisco C. Pereira, Filipe Rodrigues, Ioulia Markou, Inon Peled, K. Kaiser
Lead Organization: DTU

Description

PRISM is about designing, implementing and testing methodologies to better predict transport demand in a city. While plenty solutions exist today for this objective, there is general consensus that, under stress scenarios (e.g. large social events, inclement weather, demonstrations, special days), those approaches are insufficient. With new, smart mobility modes, demand prediction becomes even more important. For example, a one-way car sharing (e.g. DriveNow) or an autonomous mobility on demand service, are highly sensitive to rebalancing operations (moving vehicles to where demand is expected). In fact, bad demand predictions can lead to disastrous outcomes, by placing supply where it is not needed, and removing it from where it is required. PRISM approach is to combine latest research from Transport Engineering and Computer Science, by using Probabilistic Graphical Models (PGMs), a tool that combines Bayesian statistics, graph theory and scientific computing. As a research area, PGMs have already reached a considerable level of solid foundations, community size, and software tools. The Experienced Researcher (ER) has recently returned to Europe, after several years of research in Singapore and USA with the Massachusetts Institute of Technology (MIT), and this Marie SkŁodowska-Curie fellowship will be instrumental for his growth and affirmation in the Danish and European context.

Sponsor

• Marie Sklodowska-Curie Individual Fellowship H2020-MSCA-IF- 2016, ID number 745673