Glass Lizard Optimization (GLO): A Novel Nature-inspired Metaheuristic Algorithm for Engineering Applications

In this paper, a new metaheuristic algorithm called Glass Lizard Optimization (GLO) is introduced, inspired by the movement behaviors and hunting strategies of the glass lizard in nature.The design of GLO is based on two distinctive behaviors of this reptile: long-range creeping movements with spiral trajectories for extensive exploration and discovery of unknown regions, and short, focused, oscillatory movements for precise improvement of solutions around promising points.The initial population of the algorithm is generated randomly, with each member representing a candidate solution.In each iteration, the positions of the members are updated using dedicated equations from the exploration and exploitation phases, such that the exploration phase, through long-range and twisting movements, enables wide search coverage and prevents premature convergence, while the exploitation phase, with short and oscillatory movements, facilitates precise solution refinement and effective convergence.The performance of GLO was evaluated on 22 constrained optimization problems from the CEC 2011 benchmark suite and compared with nine well-known metaheuristic algorithms.The results show that GLO achieved the best mean solution in 19 out of 22 problems, demonstrating stable, competitive, and reliable performance.The Wilcoxon statistical analysis confirms the significant superiority of GLO over the competing methods.These findings indicate that, by intelligently combining the exploration and exploitation phases and leveraging the behavioral model of the lizard, GLO serves as an effective and dependable tool for solving complex constrained optimization problems in real-world applications.

Hali tarjima qilinmagan