Lin Fan

Lin Fan

Postdoc Scientist at Amazon, Incoming Prof at Northwestern

About Me

I’m a Postdoctoral Scientist in Supply Chain Optimization Technologies at Amazon, under the supervision of Garrett van Ryzin. I’m also mentored by Assaf Zeevi and Hongseok Namkoong.

I recently completed my PhD in Management Science and Engineering at Stanford University, where I was fortunate to be advised by Peter Glynn.

Starting Fall 2024, I’ll be an Assistant Professor in the Operations Department of the Kellogg School of Management at Northwestern University.

Email: linfanf [at] amazon.com

Curriculum Vitae

My research interests lie broadly at the interface of data-driven operations research and machine learning, with specializations in (1) sequential decision-making under uncertainty, and (2) statistical inference for stochastic processes arising in operations settings.

For sequential decision-making problems, I am interested in better understanding the trade-off between optimizing for average performance vs ensuring reliability/safety. My goal here is to design algorithms with good practical performance (as measured by regret), for which we more fully understand operational risks. Along this direction, I have been focused on the distribution (and especially the distributional tail properties) of regret, which provides new perspectives on algorithm design different from the conventional focus on minimizing regret in expectation.

Another thread of my work is in developing statistically and computationally efficient tools to better quantify uncertainty in stochastic system performance. I am particularly interested in flexible/robust statistical formulations of stochastic process models arising in, for example, queueing systems, where the focus has traditionally been on simpler, analytically tractable formulations. My work here also intersects with stochastic simulation and time series analysis.

Recent Talks

INFORMS 2022 Annual Meeting at Indianapolis