AI-enabled Transportation Network Analysis, Planning, and Operations

AI-enabled Transportation Network Analysis, Planning, and Operations

Yafeng Yin Headshot. The link directs to their bio page on the CCAT website.
Yafeng Yin
University of Michigan Logo - link directs to the U-M research page on the CCAT website.

Principal Investigator(s):

Yafeng Yin, Professor of Civil and Environmental Engineering – The University of Michigan
Professor of Industrial and Operations Engineering – The University of Michigan

Project Abstract:
Vehicle connectivity and automation would make vehicle trajectory data more readily available. The proposed research aims to leverage this dataset and recent advancements in implicit deep learning to develop an end-to-end modeling framework that would transform the way how metropolitan planning organizations (MPO) analyze, plan and manage their transportation networks. The proposed framework can directly take empirical, sampled trajectory data as inputs to learn drivers’ route choice behaviors and estimate traffic flow distribution across an urban traffic network. The proposed framework can further prescribe strategies such as lane direction configuration, parking provision, cordon pricing and perimeter control, to better manage the existing supply of urban traffic networks to reduce congestion.

Institution(s): University of Michigan – Ann Arbor

Award Year: 2022

Research Thrust(s): Control & Operations, Modeling & Implementation, Policy & Planning

Project Form(s):
Project Information Form