Optimizing Drone Delivery Networks Using Artificial Bee Colony Algorithm

This research investigates the Artificial Bee Colony (ABC) algorithm for optimizing drone delivery networks in urban environments, comparing it with traditional path-finding approaches such as Dijkstra's and A-star algorithms. Through experimental implementation across four strategic delivery hubs in Chandigarh, for 110 origindestination pairs and it has been found that the ABC algorithm achieved superior route optimization in 53.6% of test cases (shorter route length), with an average route distance of 3029.22 meters compared to 3052.22 meters for alternative approaches. However, this modest optimization advantage (approximately, 2530 metres for the chosen pairs) comes with significantly higher computational costs, as ABC required an average execution time of 140.41 seconds (about 2.8 times slower) versus just 36.43 seconds for Dijkstra's algorithm. Also with an optimised drone population the delivery times were reduced by 33.14 %. The study reveals a fundamental trade-off between optimization quality and computational efficiency in drone delivery networks, suggesting that while ABC's ability to balance exploration and exploitation phases makes it well-suited for complex, multi-constraint delivery scenarios, implementation decisions must carefully weigh the benefits of improved routes against increased computational demands, particularly in resource-constrained drone systems with limited onboard processing capabilities.