Learning Data-driven Transportation Models for Assessing the Effects of New Infrastructures and Rare Events

Description

The research project aims to develop new data-driven models for transportation simulation based on machine learning techniques. More specifically, the approach aims at defining multimodal transportation models over urban areas, covering modalities like personal vehicles, public transit, biking (including sharing systems) and walking, over which transportation flows is obtained from various sources of data (e.g., origin-destination surveys, traffic volume at intersections, public transit use). The model will be produced by applying some machine learning approach to extract decision-making rules over the travelling routes and modalities of users, aiming at building statistical models that are capturing behaviours found in real data. This models will be developed with the specific aim of allowing to simulate reliably the effects of changes to the transportation network (e.g., new public transit lines, urban densification, strong environmental policies incentivizing public transits), as well to the occurrence of rare events (e.g., natural disasters, major infrastructure failure). The selected candidate will work in a multidisciplinary collaboration between a research group specialized in machine learning and a group working in the field of transportation.

Requirements and Conditions

Applicants for this project must demonstrate a willingness to take part in an ambitious research project at the interface of machine learning and transportation, and a desire to establish themselves in the international academic research community.

Transmit the scientific publications or other productions demonstrating research experience.