A metaheuristic approach for optimizing drone routing in healthcare supply chains

Healthcare logistics continue to encounter significant challenges in the timely and reliable delivery of blood bags, mainly due to urban traffic congestion, rugged terrain, and the perishability of medical supplies. Conventional transportation systems frequently fall short of meeting the stringent temporal and thermal requirements inherent to healthcare supply chains. Unmanned Aerial Vehicles (UAVs), or drones, offer a compelling alternative; however, their effective deployment is hindered by constraints such as limited payload capacity, restricted flight range, narrow delivery time windows, and evolving regulatory frameworks. This study proposes the HybridNGS algorithm, a hybrid metaheuristic framework that integrates Nearest Neighbour (NN) for solution initialization, Genetic Algorithm (GA) for global search, and Simulated Annealing (SA) for local refinement, to address the Drone Routing Problem (DRP) in blood logistics. The model incorporates domain-specific constraints, including blood-type compatibility, energy-aware routing, and cold-chain preservation. Empirical evaluations using synthetic and real-world datasets comprising 20–100 hospitals reveal that HybridNGS consistently outperforms benchmark approaches such as GRASP and TSP-D, achieving up to 20 % cost savings, 15 % reduction in drone usage, and notable energy efficiency. The algorithm demonstrates strong scalability and robustness under variable demand and environmental conditions. It is a viable solution for enhancing accessibility, reliability, and sustainability in routine and emergency healthcare delivery systems. • Propose a hybrid method to optimize drone routes for medical supply delivery under real-world constraints. • Develop a model that captures payload, energy, and thermal safety conditions in blood transport logistics. • Introduce adaptive techniques to balance cost, feasibility, and exploration in hospital network routing. • Demonstrate consistent performance gains in delivery cost and resource use across diverse hospital systems. • Bridge a gap in supply chain analytics by modeling realistic medical drone delivery scenarios.