Brief Introduction
The workshop will deliver cutting‑edge insights into operations management and data‑driven decision‑making research, and support the creation of high‑quality papers.

Time and Venue
Date: May 14, 2026 (Thursday)
Venue: E22-G015

Organizing Committee
Prof. Grace Qi FU, Prof. Li XIAO

Workshop Agenda

*UM Distinguished Visiting Scholar of UM Talent Programme 

Speakers and Talks Information

Prof. Sean Xiang ZHOU

Prof. Sean ZHOU is currently a Professor and the Chairman in Department of Decisions, Operations and Technology, The Chinese University of Hong Kong (CUHK) Business School. He is also the Director of the Supply Chain Research Centre under the Asian Institute of Supply Chain and Logistics in CUHK. Prior to joining the Business School, he was a faculty member at Department of Systems Engineering and Engineering Management, CUHK. He received his BS in Electrical Engineering from Zhejiang University, China in 2001 and his MS and PhD in Operations Research from North Carolina State University in 2002 and 2006, respectively. His main research area is supply chain management with particular interests in inventory control, production planning, dynamic pricing, and game theoretic applications. He is an Area Editor of OR Letters, Senior Editor of POM and Associate Editors of JORSC, JSSSE, NRL, Service Science.

Topic: Optimal Growth of a Two-Sided Platform with Heterogeneous Agents

Abstract: We consider the dynamics of a two-sided platform, where the agent population on both sides experiences growth over time with heterogeneous growth rates. The compatibility between buyers and sellers is captured by a bipartite network. The platform sets commissions to optimize its total profit over T periods, considering the trade-off between short-term profit and growth as well as the spatial imbalances in supply and demand. We design an asymptotically optimal policy with the profit loss upper-bounded by a constant independent of T, in contrast with a myopic policy shown to be arbitrarily bad. To derive the policy, we first develop a benchmark problem that captures the platform’s optimal steady state. We then identify the agent types with the lowest relative population ratio compared to the benchmark in each period, and adjust the service level of these types to be higher than or equal to their service level in the benchmark problem. A higher service level accelerates growth but requires substantial subsidies during the growth phase. Additionally, we provide the conditions under which the subsidy is necessary. We further examine the impact of the growth potential and the compatibility network structure on the platform’s optimal profit, the agents’ payment/income, and the optimal commissions at the optimal steady state. To achieve that, we introduce novel metrics to quantify the long-run growth potential of each agent type. Using these metrics, we show that a “balanced” compatibility network, where the relative long-run growth potential between sellers and buyers for all submarkets is the same as that for the entire market, allows the platform to achieve maximum profitability.  Our study provides insight into how the growth potential and compatibility network structure jointly influence the commission policy in the growth process and the optimal steady state. This is joint work with Yixin Zhu (SUFE), Kevin Chen (CUHK) and Philip Zhang (CUHK).

Prof. Kairen ZHANG

Prof. Kairen ZHANG is an Associate Professor in the Department of Logistics Management Engineering at Southeast University. He received his Ph.D. degree in systems engineering & engineering management from Department of Systems Engineering & Engineering Management at the Chinese University of Hong Kong in 2016. He also holds a M.S. in management science from Fudan University and a B.S. in mathematics and applied mathematics from Hunan University. His current research interests focus on data-driven inventory control, supply chain management, project management.

Topic: A Hybrid Sampling-Based and Gradient Descent: Learning Method and Its Applications

Abstract: We develop an algorithmic framework for sequential decision-making problems facing unknown uncertainty over multi-periods. In particular, the decision maker (DM) needs to make two decisions each period while learning from the previous realized uncertainty. The framework integrates sample average approximation (SAA) for the first-stage decision with stochastic gradient descent (SGD) for the second-stage decision, and the resulting generic algorithm attains an O(√T) regret bound. We apply and adapt the framework to develop the algorithms for both the repeated settings with applications such as the capacity allocation and final-buy problems where each period the problem resets itself and the dynamic setting with applications such as multi-echelon inventory systems and dual-sourcing problems where state variable(s) couple the periods together. For the latter class of problems, we overcome the technical challenges arising from the impact of decisions propagates through the state variables that affect both the feasible set and cost structures to establish the regret bounds of the developed algorithms. This is achieved by constructing auxiliary functions that help bound the resulting additional component in the overall regret. Managerial Implications: The algorithms developed for various applications demonstrate great performance numerically and outperform several existing ones. When generalized to the case with more than two decisions each period (under some additional assumption), our framework allows flexible allocation of SAA and SGD components, enabling the DM to adaptively manage data and computational efficiency.

 Prof. Li XIAO

Prof. Li XIAO is an Associate Professor at the Faculty of Business Administration, University of Macau. Her research primarily focuses on supply chain management and service operations. She earned her Ph.D. from the National University of Singapore, an M.Sc. from the Chinese Academy of Sciences, and a B.S. from Wuhan University. Prior to joining the University of Macau, she held academic positions including Postdoctoral Fellow and Research Assistant Professor at the Chinese University of Hong Kong, Assistant Professor at Tsinghua University, and Associate Professor at the Southern University of Science and Technology. Her work has been published in leading journals such as Operations Research and Production and Operations Management.

Topic: Learning to Order under Substitution: Online Gradient Descent with Cyclic Exploration

Abstract: We consider online inventory control with lost sales and stockout substitution. The setting presents two hurdles: censored observations and an unidentifiable portion of sales arising from substitution. We propose a learning framework that couples cyclic exploration with minibatch gradient descent, which is flexible enough to accommodate multiple information regimes. In the sales-only case, we prove the algorithm’s average cost converges to the clairvoyant benchmark.