Customer-to-Customer Delivery: A Compensation-Driven Incentive for Customers to Take on Delivery Orders

• Applicant: Professor Dr. Rouven Schur
• Subject Area: Operations Management and Computer Science for Business Administration
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 550416070

Online commerce has been rapidly growing for several years, and with it, the need to efficiently manage last-mile delivery. To achieve this, both academia and industry are exploring innovative delivery approaches, such as Crowdsourced Delivery, where customers deliver to other customers. As an illustrative example, consider the grocery industry: some customers order groceries online while others shop in-store. Those shopping in-store can, after their purchase, deliver orders to other customers on their way to their destination (e.g., home or workplace). These delivering customers, referred to as occasional drivers, receive compensation to offset the detour required for delivery. The company’s aim is to deploy occasional drivers in a way that minimizes delivery costs. Previous research on Crowdsourced Delivery often assumes that all occasional drivers are constantly available and will always accept assigned tasks (as long as pre-established conditions are met), with compensation schemes considered fixed. These assumptions typically lead to deterministic and static optimization models. However, it is more realistic to assume a dynamic and stochastic setting: individual occasional drivers are not always available, and their willingness to take a detour for a certain compensation is not known in advance. In this more complex scenario, the order-specific and driver-specific adjustment of compensation becomes a critical tool for incentivization. Although aspects such as uncertainty, dynamic decisions, or setting compensation levels have received attention in research, there is a lack of integrated approaches that combine these elements into a comprehensive model to accurately depict Crowdsourced Delivery. This project aims to fill this research gap. It examines three stochastic settings, two of which are also dynamic. In each setting, the optimal order-specific compensation offers are determined, actively considering the stochastic decisions of occasional drivers and the influence of compensation height on these decisions. The companies’ goal is always to minimize delivery costs. In the two dynamic settings, occasional drivers arrive spread over the planning period (e.g., a business day), allowing the company to dynamically adjust compensation levels according to the current situation. In one of the dynamic settings, it is additionally assumed that the destinations of occasional drivers are known in advance, enabling the company to select compensation levels that are both order-specific and driver-specific.

• DFG Programme: Research Grants
• Co-Investigator: Dr. Matthias Soppert