A modified Ant Colony Optimization algorithm to solve a dynamic traveling salesman problem: A case study with drones for wildlife surveillance

Abstract

Abstract This study presents a novel Ant Colony Optimization (ACO) framework to solve a dynamic traveling salesman problem. To maintain diversity via transferring knowledge to the pheromone trails from previous environments, Adaptive Large Neighborhood Search (ALNS) based immigrant schemes have been developed and compared with existing ACO-based immigrant schemes available in the literature. Numerical results indicate that the proposed immigrant schemes can handle dynamic environments efficiently compared to other immigrant-based ACOs. Finally, a real life case study for wildlife surveillance (specifically, deer) by drones has been developed and solved using the proposed algorithm. Results indicate that the drone service capabilities can be significantly impacted when the dynamicity of deer are taken into consideration. Highlights Proposed a novel ACO-ALNS based metaheuristic. Four variants of the proposed metaheuristic is developed to investigate the efficiency of each of them. A real life case study mirroring the behavior of DTSP is developed.