创造管理新知,造就业界精英,践行持续发展,应对全球挑战。
Our Mission
【2026年12月2日】【管理高等研究院公开学术报告-第167期】基于随机化的鲁棒定价:多准则视角
Speaker: Dr. Mingyang Fu, University of Toronto
Date & Time: Wednesday, 2 December 2026, from 14:00 to 15:30 AM (Beijing Time)
Location: Tongji Building A2101
ABSTRACT
This paper addresses a data-driven decision-making problem in which a seller seeks to optimize the price of a single item for a buyer, using historical data that includes observed price-and-conversion-rate pairs. Given demand uncertainty, the goal is to design a pricing strategy that is robust across multiple criteria in the worst-case scenario. We establish a simple and easily verifiable necessary and sufficient condition for the existence of a randomized pricing mechanism that satisfies given performance targets across three key robust criteria: worst-case expected revenue, worst-case ratio, and worst-case regret. We propose a class of randomized pricing mechanisms, parameterized by only three values, that simultaneously balance all three metrics and achieve any attainable performance target. In addition, we characterize the Pareto frontier for these three robust criteria, providing insights into the trade-offs among them and offering valuable guidance for sellers facing pricing decisions under uncertainty. Our analysis also demonstrates that even slight randomization, with a modest sacrifice in worst-case revenue, can significantly improve robustness regarding ratio and regret. Furthermore, we identify the optimal mechanisms for each robust criterion by solving a one-dimensional convex optimization.
Keywords: pricing, mechanism design, data-driven decision making, distributionally robust optimization, limited information, multi-criteria
GUEST BIO
Mingyang Fu is a Postdoctoral Research Fellow at the Rotman School of Management, University of Toronto. He earned his Ph.D. in Industrial Systems Engineering from the National University of Singapore and his BSc in Statistics from Nanjing University. His research focuses on distributionally robust optimization, pricing and mechanism design, and supply‑chain analytics. His work has appeared in Manufacturing & Service Operations Management and an EC conference, with multiple manuscripts under major revision at top‑tier OR‑MS journals. He also serves as an ad hoc reviewer for leading operations management journals.