Joint order matching, routing and compensation pricing in crowdsourced pickup and delivery

Abstract

This paper considers a crowdsourced delivery platform that integrates both a professional fleet and occasional drivers (ODs) to fulfill customer orders. The delivery willingness of ODs is measured by their maximum tolerable detour time, which is influenced by the compensation price they receive by a compensation-willingness function. Our objective is to minimize total delivery cost by developing an integrated order matching, routing and compensation pricing optimization model for crowdsourced pickup and delivery, in which the compensation price is treated as a decision variable and the compensation-willingness function is incorporated as a behavioral constraint for ODs. To solve this problem, we propose a hybrid algorithm combining an adaptive golden-section search method with adaptive large neighborhood search. Numerical experiments demonstrate that the proposed algorithm is competitive in both solution quality and computation time compared with CPLEX and a multi-start heuristic combining variable neighborhood search and tabu search. A case study indicates that the optimized compensation price scheme maintains ODs’ willingness to perform deliveries and reduces the total delivery cost by approximately 23.5% compared with a constant low compensation price scheme and by about 9.0% compared with a constant high compensation price scheme, thereby effectively preventing resource waste and insufficient incentives