Improved Ant Colony Algorithms for Eliminating Stagnation and Local Optimum Problem — A Survey

ACO (Ant Colony Optimization) is popular swarm intelligence with stochastic nature meta-heuristic algorithm applied to solve many combinatorial optimization problems. The characteristics of ACO includes robust, positive feedback, distributed computing and easy fusing with other algorithms makes ACO simpler and efficient in searching optimal solutions. But the ACO algorithm has various drawbacks like stagnation, low convergence speed, local optimum problem and long search time (computation time). These problems in ACO can be reduced or eliminated by improving various parameters like pheromone updating factor, pheromone evaporation factor, initial distribution of pheromone value, etc. The search process can be induced by integrating ACO with other algorithms like mutation operators, crossovers of GA (Genetic Algorithm) and PSO (Particle Swarm Optimization), Acoustic factors, ABC (Artificial Bee Colony). This paper gives the survey of all the techniques added to the traditional ACO algorithm to improve its performance by removing all the above mentioned problems along with the summarization. Finally the idea of parameter tuning based on the application requirements is discussed.