Torres, F., Gendreau, M., and Rei, W. (2024)

Pricing Routes: Compensation of Crowd-Drivers

Compensating crowd-drivers (CD) to complete delivery tasks requires careful analysis. Crowd-drivers are on-demand independent workers that can reject undesirable delivery requests. In companies like Uber, pricing delivery tasks is important to adjust the supply of Uber drivers to match the demand. Unlike Uber, e-commerce companies (e.g., AmazonFlex) have a set of delivery requests that are more cost efficient when consolidating multiple delivery requests in a single route. The compensation influences the probability of route acceptance. As a consequence, route optimization and pricing need to be integrated together to minimize the expected cost and maximize the participation of CDs in a crowd-shipping platform. In this study, we develop a framework to create an operational plan of routes, and price routes based on the market. The probability of route acceptance by CDs is modeled with an econometric model. The model considers different route attributes that influence the probability of acceptance (e.g., distance, location, compensation, load and number of stops). The operational problem is a variant of the Heterogeneous Vehicle Routing Problem where a fleet of commercial vehicles and a pool of stochastic CDs can complete delivery requests. We develop an Adaptive Large Neighborhood Search (ALNS) algorithm to integrate all decision in a single framework. In addition, we investigate strategic objectives for the long-term success and sustainability of the platform and provide managerial insights. In particular, our study demonstrates three main results: 1) A fleet of only crowd-drivers is more robust than a fleet of commercial vehicles for the various demand scenarios 2) The compensation of CDs remains stable while incrementing participation and 3) A large increase of participation can be achieved with a small increase of the expected cost of deliveries. Our analysis offers insights for effectively pricing multiple on-demand delivery requests with independent workers.