On-Demand Service Sharing via Collective Dynamic Pricing

• 成果类型: Article
• 署名作者: Dogan, Mustafa; Jacquillat, Alexandre
• 署名单位: University of East Anglia; Massachusetts Institute of Technology (MIT)
• 刊物名称: M&SOM-MANUFACTURING & SERVICE OPERATIONS MANAGEMENT
• ISSN/ISSBN: 1523-4614
• DOI: 10.1287/msom.2024.1301
• 发表日期: 2026
• 关键词: lead-time quotation; mechanism design; queuing-systems; public-goods

摘要

Problem definition: This paper studies an on-demand service sharing problem, motivated by emerging operating models in ride-sharing, food delivery, and made-toorder manufacturing. Time-sensitive customers arrive dynamically onto a platform with heterogenous willingness to pay and private information. The platform can serve each customer individually or pool customers together, giving rise to interdependencies between customers and over time. This goal is to optimize who to serve, when, and at what price. Methodology/results: We formulate a dynamic allocation and pricing mechanism to maximize the platform's expected discounted profits, subject to incentive compatibility and individual rationality constraints. We prove that the problem can be decomposed via dynamic programming, based on the novel notion of collective virtual value, defined as the marginal revenue that the platform can extract from all customers. The optimal mechanism follows a simple, easily implementable index rule: service is provided whenever the collective virtual value exceeds a threshold that decreases with the number of available suppliers. Managerial implications: Service sharing enables temporal discrimination: the platform provides immediate or delayed services based on customers' own willingness to pay, but also on the time of their requests and demand from other customers. In practice, on-demand service sharing can be managed via a dynamic menu to offer differentiated service levels and prices, trading off cost-minimization, demand-supply management, and discriminatory objectives. Our results show that even simple dynamic menus can outperform benchmarks based on posted prices and can lead to win-win outcomes for the platform and consumers.