Yeonu Chae

I am a Ph.D. student in Industrial and Operations Engineering at the University of Michigan, advised by Prof. Siqian Shen. Before that, I received B.S. degrees in Industrial Engineering and Artificial Intelligence from Seoul National University.

My research focuses on optimization under uncertainty for shared mobility systems, particularly electric mobility and autonomous transportation. I am interested in how uncertainty in demand, technology, and human behavior shapes strategic planning and operational decisions in these systems.

Interests Robust optimization · Stochastic programming · Shared mobility · Electric vehicle fleets · Battery degradation

Portrait of Yeonu Chae

Education

Honors and Awards

Research

Shared fleets are large, expensive, and exposed to uncertainty in demand, energy, and vehicle condition. My research develops optimization models that make planning and operating decisions together, so that what is built matches how it will actually be run.

Battery degradation-aware planning and operations of shared EV fleets

Electric vehicles lose usable capacity as their batteries age, and how a fleet is charged and dispatched changes how fast that happens. I model fleet sizing, charging, and dispatch jointly with degradation, and use a robust satisficing framework that targets an acceptable outcome across uncertainty instead of optimizing only for the expected or worst case. With Mengshi Lu and Siqian Shen.

Infrastructure and operations co-design for mixed autonomous and human-driven ride-hailing

Autonomous vehicles are entering ride-hailing fleets alongside human drivers, and the two have different costs, ranges, and constraints. I study how to design infrastructure and operating policies together for these mixed fleets under demand uncertainty, so that investment decisions reflect the operations they will support. With Siqian Shen and Mengshi Lu.

Data-driven battery health prediction

At Seoul National University, I applied time-series deep learning models such as FEDformer, TimesNet, and the non-stationary Transformer to battery remaining-useful-life prediction, improving prediction R² by roughly 110% through feature selection. This work is the data side of the degradation question my current research addresses with optimization. With Prof. Taesup Moon's group.

Depot Charging station Demand node Schematic of a shared-fleet network, not data from a study.

Publications

Working papers

  1. Battery Degradation-aware Planning and Operations of Shared Electric Vehicle Fleets with Robust Satisficing. Yeonu Chae, Mengshi Lu, and Siqian Shen. To be submitted, 2026.
  2. Infrastructure and Operations Co-Design for Mixed Autonomous and Human-Driven Ride-Hailing Fleets. Yeonu Chae, Siqian Shen, and Mengshi Lu. To be submitted, 2026.

In preparation

  1. A Unified Deep Learning Framework for User-Driven Battery RUL Prediction. Yong Jin Jeong, Hyunjae Kim, Sangwon Jung, Seongha An, Yeonu Chae, Gahyun Lee, Inwoo Kim, Jun Hee Lee, Chun Yong Kang, and Jang Wook Choi. In preparation for Energy & Environmental Science.

Talks

Experience

Teaching & Service

Teaching

  • Head Teaching Assistant, AI Big Data Camp Institute for Industrial Systems Innovation, Seoul National University · Aug 2023 and Jan 2024

    Organized two camps for more than 80 high school students and coordinated ten teaching assistants across lectures and hands-on sessions.

  • Peer Tutor in Industrial Engineering Seoul National University and Nobel Leadership Camp · 2021–2024

    Tutored operations research and statistics: Markov chains, Poisson processes, renewal theory, estimation, hypothesis testing, and regression.

Service and leadership

  • Vice President, Korean Graduate Student Council University of Michigan
  • President, AttentionX AI Research Club Seoul National University
  • Vice President, Industrial Engineering Student Council Seoul National University

Curriculum Vitae

The full CV, including coursework and technical skills, is available as a PDF.

Download CV (PDF)

Last updated September 2026.