Lagrangian decomposition heuristic for an inventory and distribution planning problem with unmanned aerial vehicles in a blood supply chain

초록

This study examines the joint inventory and distribution planning problem for a blood supply chain with a blood centre and hospitals, inspired by the practices of the Korean Red Cross. Blood products are distributed using conventional blood vans and unmanned aerial vehicles, commonly known as drones. The problem is to determine the inventory lot size, distribution schedule, and transport mode (vans and drones) while satisfying the demand for blood products at hospitals over a discrete planning horizon. The objective is to minimise the costs associated with blood processing, transportation, and inventory holding at the blood centre and hospitals. To address this, the problem is initially formulated as a conventional lot-sizing model and then transformed into a lot-sizing model in the form of a transportation problem. A Lagrangian decomposition heuristic algorithm is proposed to decompose the original problem into smaller subproblems for more efficient solving. The optimal properties for the decomposed subproblems are analytically derived. Computational experiments using real-world data demonstrate that the proposed heuristic generates near-optimal solutions within a short computation time. Furthermore, sensitivity analysis shows that drone adoption significantly improves cost savings in the blood supply chain, providing meaningful managerial insights, including the critical factors for drone adoption.

키워드