Yufeng Cao, Associate Professor from the Department of Management Science at Antai College of Economics and Management, Shanghai Jiao Tong University, has co-authored a paper published in Management Science, one of the world’s most prestigious peer-reviewed journals in management. The article was officially released in April 2026, titled Dynamic Pricing for Two-Sided Marketplaces with Offer Expiration (2026, 72(4): 3081-3100).

Paper Abstract

The research examines a two-sided service marketplace in which platform operators procure services from suppliers and resell them to end customers with differentiated dynamic pricing strategies. Within the ecosystem, services are exchanged in the form of standardized discrete tasks, whose core attributes include task category, acceptance deadline, as well as corresponding prices for customers and service providers. Customers release service tasks to the marketplace, whereas eligible vendors pick suitable assignments according to their operational preferences.

The team formulates the pricing dilemma into an infinite-horizon average-reward Markov decision process that accounts for the arrival patterns and heterogeneous decision-making behaviors across both customer and supplier sides. To address computational complexity stemming from large-scale market data and the curse of dimensionality, the study leverages a discrete-time fluid approximation method. The analytical outcome reveals a simplified yet effective pricing principle: the optimal price curve for a single task is exclusively governed by its remaining valid time before deadline, without being affected by other ongoing tasks on the platform. The proposed pricing policy is proven asymptotically optimal, maintaining a long-term performance loss ratio of O(1/θ), where the parameter θ represents the overall scale of market supply and demand. The research further benchmarks the developed pricing rule against mainstream heuristic strategies including static fixed pricing and state-dependent pricing. The base model is also expanded to adapt to real-world continuous-time scenarios and limited-operation-cycle platforms.

About the Author

Yufeng Cao

Associate Professor, Department of Management Science, Antai College of Economics and Management, Shanghai Jiao Tong University

Research Interests: Optimization modeling and algorithm design, decision analysis, statistical learning, with applications covering supply chain and revenue management, mechanism design for multi-sided platforms, and interdisciplinary research integrating operations management and artificial intelligence.