Our research lab focuses on innovating next-generation modeling and computational tools for mobility and logistics systems, with a emphasis on connectivity, electrification, and automation. Our core research directions include, but are not limited to:
1. AI-Enabled Modeling and Optimization of Transportation Networks
By 2035, nearly half of all new vehicles in the United States will be connected, generating unprecedented volumes of mobility data. We are developing an AI-enabled inverse learning framework that leverages advances in machine learning and corwdsourced mobility data to transform the foundational paradigm of transportation network equilibrium modeling — a pillar of transportation system planning and management for over seventy years.
2. Economic Modeling and Regulation in Multimodal Transportation and the Sharing Economy
Ridesourcing vehicles generate nearly three times the vehicle miles traveled compared to private vehicles and are destined to be the next special fleet for electrification. However, high purchase costs and limited access to fast charging remain major barriers. This direction develop a novel aggregate equilibrium model for the electrified ridesourcing system and design innovative (i) regulatory policies and (ii) differential matching mechanisms to accelerate the electrification.
3. Learning, Adaptation, and Policy Design in Human–Autonomous Vehicle Interactions
By 2045, half of all new vehicles sold in the U.S. are expected to be autonomous. As humans and AVs increasingly share the road, complex interaction dynamics will emerge. While some believe AVs will enhance safety, others raise concerns that AVs’ strict adherence to traffic laws may provoke aggressive behaviors — for instance, pedestrians intentionally running red lights — potentially destabilizing traffic flow. This research seeks to model and inform the co-adaptation of humans and autonomous systems. Our goal is to support policymakers in designing behavioral and regulatory strategies that improve safety, reduce congestion and emissions, and facilitate broader AV adoption.
📖 Publications
🌟 Featured Papers
End-to-End Learning of User Equilibrium: Expressivity, Generalization, and Optimization
Zhichen Liu, Yafeng Yin, Transportation science, 2025
End-to-end learning of user equilibrium with implicit neural networks
Zhichen Liu, Yafeng Yin, Fan Bai, Donald K Grimm, Transportation Research Part C: Emerging Technologies, 2023
Regulatory policies to electrify ridesourcing systems
Zhichen Liu, Zhibin Chen, Yafeng Yin, Zhengtian Xu, Transportation Research Part C: Emerging Technologies, 2022
Other Papers
- Dynamic origin-destination flow prediction using spatial-temporal graph convolution network with mobile phone data, Zhichen Liu, Zhiyuan Liu, Xiao Fu, IEEE Intelligent Transportation Systems Magazine, 2021.
- Estimating Sectional Volume of Travelers Based on Mobile Phone Data, Zhichen Liu, Xiao Fu, Yang Liu, Weiping Tong, Zhiyuan Liu Journal of Transportation Engineering, Part A: Systems, 2020.
Research in Progress
- Inverse Learning of Congestion Game via Multiconvex Optimization. Zhichen Liu, Xi Lin and Yafeng Yin. Under review with the 26th International Symposium on Transportation and Traffic Theory.
- Distributionally Robust Transportation Networks Design with Contextual Uncertainty. Zhichen Liu, Yafeng Yin, and Xi Lin. Preprint accepted at the 2025 TRB Annual Meeting.
- Electrify Ridesourcing System with Differential Matching. Preprint accepted at CASPT2025.
- Large-Scale Inverse Learning of User Equilibrium via Multiconvex Optimization. Zhichen Liu, Yafeng Yin, and Xi Lin. Under review with 26th International Symposium on Transportation and Traffic Theory.