Start of Main Content
Operations

Donald P. Jacobs Scholar

Assistant Professor of Operations

Headshot of Lin Fan, faculty at the Kellogg School of Management

Lin Fan is an Assistant Professor of Operations at the Kellogg School of Management (since September 2024). Prior to joining Kellogg, he spent one year as a Postdoctoral Scientist at Amazon in Supply Chain Optimization Technologies. He received his PhD in Management Science and Engineering, an MS in Statistics, and an MS in Mechanical Engineering, all from Stanford University. His research interests lie broadly at the interface of applied probability and data-driven operations, with specializations in multi-armed bandits, reinforcement learning, statistical inference for stochastic processes, and stochastic simulation.

About Lin
Research interests
  • applied probability + data-driven operations: multi-armed bandits
  • reinforcement learning
  • statistical inference for stochastic processes
  • and stochastic simulation
Teaching interests
  • operations management
  • sequential learning and decision-making under uncertainty
  • stochastic models
  • PhD, 2023, Operations Research, Stanford University
    MS, 2017, Statistics, Stanford University
    MS, 2015, Mechanical Engineering, Stanford University
    BS, 2012, Mechanical Engineering, Georgia Institute of Technology, Highest Honors
  • Assistant Professor, Operations, Kellogg School of Management, Northwestern University, 2024-present
    Postdoctoral Scientist, Amazon - Supply Chain Optimization Technologies, 2023-2024
  • Stanford University Centennial Teaching Assistant Award
    Second Place, George Nicholson Student Paper Competition (for the paper "The Fragility of Optimized Bandit Algorithms"), INFORMS
    National Science Foundation Graduate Research Fellowship, 3 years

Emerging Areas in Operations Managements (OPNS-525-0)

This course studies novel, emerging topics and methods used in academic research of operations management. Content will depend on the expertise and interests of the instructor. Past content included statistical (machine) learning and sequential decision-making, such as bandit learning, balancing exploration/exploitation, and reinforcement learning, including methods for value function approximation and algorithms for efficient exploration.