Crowd-shipping delivery performance from bidding to delivering

Crowd-shipping is an innovative delivery model using digital platforms to match the demand for shipments with supply using excess transport capacity and drivers from the crowd. This sharing economy delivery concept has attracted growing attention to address the pressing challenges of urban goods deliveries. Little is known about the actual performance of crowd-shipping platforms due to limited data-availability and operational transparency. A particular challenge is that part of the delivery outcome is determined in the platform's digital space related to bidding and matching of supply and demand, followed by a real-world delivery operation, typically carried out by non-expert couriers. This paper provides the first comprehensive analysis of the entire crowd-shipping process from the bidding stage, through shipment acceptance, pickup, and final delivery. Using parametric hazard modeling applied to a unique U.S. national database of 16,850 crowd-shipping delivery instances, the authors examine which factors play a role in each phase of the delivery process. The findings illustrate that shipping requests and packages, built environment, and socioeconomic characteristics have a variable impact on each delivery stage. In particular, posting in the morning or evening hours and for business-to-consumer shipments significantly accelerates the digital phase, but has no effects on the final delivery phase. Moreover, the results reveal that performance loss occurs non-uniformly in the platform process, with a more significant loss in delivery rates related to the digital posting and bidding. A more substantial loss of delivery speed performance occurs in converting from digital to real delivery in negotiating the pickup arrangement. Crowd-shipping companies will benefit from the research to improve the management of their peer-to-peer-based mechanism.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01762691
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Dec 23 2020 3:07PM