Moth-Flame Optimization and Ant Nesting Algorithm: A Systematic Evaluation
Hanan K. AbdulKarim, Tarik Ahmed Rashid · 2023
In this paper, some swarm metaheuristic algorithms are analyzed to show the performance of algorithms.Swarm Intelligence algorithms are more trapping in local optima because of low exploration.The Moth-Flame Optimization algorithm is a widely applied metaheuristic algorithm.Nevertheless, the Ant Nesting Algorithm is another recent powerful algorithm.Both algorithms are theoretically and practically studied and applied to a simple optimization problem with a simple objective function.All steps of the algorithms are implemented and discussed providing results to show and compare the exploration and exploitation performance between both algorithms.A simple comparison between both algorithms is conducted using the same sample of data, and it is concluded that convergence within cycles of implementation shows that Ant Nesting Algorithm is fast converged but it might get stuck in local optima because of low exploration.Additionally, the merits of the Ant Nesting Algorithm show the algorithm will stuck in local optima when solving highly complex problems or if the initial population can't be explored efficiently.Moth-Flame Optimization also may have exploration problems when the size of flames during cycles is decreased.