A review of exploring recent advances in ant colony optimization: applications and improvements

Abstract

Inspired by the foraging behavior of ants, the well-known metaheuristic Ant Colony Optimization ‎‎(ACO) provides strong answers to challenging optimization issues in many spheres. This work ‎investigates current developments in ACO algorithms with an emphasis on hybridization, employing methods including machine learning, adaptive mechanisms, and genetic algorithms to ‎improve performance. Applications such as robotics, telecommunications, healthcare, and logistics ‎show ACO's adaptability in handling path planning, resource allocation, and data optimization. ‎Dynamic pheromone methods, multi-objective optimization, and domain-specific adaptations ‎, which have raised computing efficiency, scalability, and solution quality, have been key advances. ‎Notwithstanding these developments, problems, including parameter sensitivity and real-time ‎adaptation, remain unresolved. Future studies include integrating real-time data, creating scalable ‎adaptive algorithms, and tackling domain-specific restrictions to further increase ACO's relevance. ‎This work emphasizes ACO's possible importance as a fundamental instrument for addressing ‎problems of real-world optimization‎.