Data-Driven Decision Making in Dispatch and Location Models

Date: 2022/07/22 - 2022/07/22

Academic Seminar: Data-Driven Decision Making in Dispatch and Location Models

Speaker: Cheng Hua, Assistant Professor, Department of Management Science, Antai College of Economics and Management, Shanghai Jiao Tong University

Time: 10:00 - 11:30, July 22, 2022 (Beijing Time)

Location: via Feishu

Abstract

The talk includes two topics: optimal dispatch by reinforcement learning and optimal unit locations using Bayesian Optimization. The first topic models the dispatch problem as an average-reward Markov decision process to find an optimal dispatch policy. We propose a temporal difference learning (TD-learning) approach that approximates the state-value of the optimal scheduling problem based on the post-decision state using domain knowledge. The second topic proposes a Bayesian optimization solution to the optimal location problem that includes searching within feasible trust regions and adaptive swapping strategies. We show that our algorithm always converges to a globally optimal solution with a regret-bound guarantee. We also apply our methods to a case study in St. Paul, Minnesota, using one year of real data.

Biography

Cheng Hua is an Assistant Professor in the Department of Management Science at the Antai College of Economics and Management, Shanghai Jiao Tong University. He received his Ph.D. in Operations Research from the Yale School of Management. He also received two master's degrees from Yale University and bachelor's degrees from Shanghai Jiao Tong University and the University of Michigan, Ann Arbor. He has received 7 international best paper awards in operations research and management science. He also has industry experience at Apple, Pfizer, and Deloitte. His research interests are data-driven decision-making, healthcare operations management, AI for Science, and quantitative trading.