465. Dynamic Compensation Optimization for Occasional Driver Delivery
Invited abstract in session MB-29: Optimization in Mobility and on-demand services, stream Pricing and Revenue Management Innovations.
Monday, 10:30-12:00Room: Maurice Keyworth 1.04
Authors (first author is the speaker)
| 1. | Kai Winheller
|
| Universität Duisburg-Essen | |
| 2. | Matthias Soppert
|
| Chair of Business Analytics & Management Science, University of the Bundeswehr Munich | |
| 3. | Rouven Schur
|
| University of Duisburg-Essen |
Abstract
Despite considerable scientific efforts and despite the substantial advancements made in the literature, the last-mile remains the supply chain's most costly component. In this context, the concept of crowd-engagement in last-mile delivery recently gained attention also for deliveries from brick-and-mortar stores. However, existing approaches assume that the store possesses detailed information on the crowd or that it possesses the control ability to assign delivery tasks to the crowd.
In our work, we relax both of these two key assumptions. That is, on the one hand, we assume that the pool of potential drivers as well as their specific preference profiles are unknown to the store. On the other hand, the store's ability to engage the crowd is limited in the sense that compensations can be optimized dynamically but that a specific driver cannot be engaged with certainty. Thus, we consider occasional drivers in a narrower sense, where in-store customers may occasionally become drivers who deliver to online customers, depending on the compensations offered and the required detours. We formulate the problem as Markov decision process. Based on analytical analyses of the problem and the derivation of the exact solution for an important special case, we develop scalable heuristic solution approaches.
Keywords
- Dynamical Systems
- Stochastic Models
- Transportation
Status: accepted
Back to the list of papers