Zhichen Liu is a Ph.D. candidate in Civil Engineering at the University of Michigan under the supervision of Professor Yafeng Yin. Her research specializes in AI-enabled transportation network analysis and her works have been published in leading journals. Zhichen is on the board of the Michigan Transportation Student Organization. She serves as a reviewer for top-tier journals and for the Transportation Research Board Standing Committee. Her efforts have been acknowledged with an Engineering Graduate Fellowship from the University of Michigan.
Traditional equilibrium analysis frameworks, established in the 1950s, rely heavily on the modeler's judgment and lack empirical validation. Zhichen's research seeks to revolutionize the transportation networks analysis process with an end-to-end framework that directly learns transportation network equilibrium models from multi-source data; it melds the model-building and policy-optimization stages, which were separated in the traditional framework, into a single end-to-end streamline. The proposed learning algorithm is robust with theoretical performance guarantees and is being validated with real-world network-level connected vehicle trajectory data from Ann Arbor, Michigan.