Two-echelon collaborative routing problem with heterogeneous crowd-shippers

Abstract

In this study, we consider a setting in which a retail store uses occasional couriers (OCs), which are not regular couriers but work as couriers on occasion, and in-store customers, which make a single delivery on the way back to their origin locations from the retail store, to perform the last-mile delivery along with its own fleet of couriers. To represent this setting, we design a two-echelon hetero-collaborative routing problem (2E-HCRP) that facilitates parcel transfers between couriers and OCs at customer nodes, in addition to in-store customer deliveries for optimizing last-mile services. We develop a mixed-integer linear programming (MILP) model and an adaptive large neighborhood search (ALNS) heuristic algorithm to devise cost-effective delivery plans that incorporate couriers, OCs, and in-store customers simultaneously. Furthermore, we introduce novel neighborhood search algorithms specifically tailored for parcel transfers and a scheduling algorithm designed for enhancing OC utilization and timely delivery to customers. We evaluate the performance of the proposed algorithms with the genetic algorithm, Monte Carlo method, and neighborhood search. The results indicate that our proposed algorithms outperform the others in minimizing travel costs and time window violations. Moreover, we provide valuable managerial implications through sensitivity analysis.