Dynamic evaluation of Decomposition Methods for Large-Scale Optimization Problems using an Island Model

Optimization problems with a high number of variables are known as Large-Scale Optimization Problems (LSOPs) and tend to be complex to solve. Additional strategies can be applied in Evolutionary Algorithms (EAs) to solve LSOPs. Decomposition Methods (DMs) decompose the problem domain into groups, then solve them separately. This work implements an adaptive hybrid Island Model based on stigmergy to solve LSOPs using different DMs. The DMs are compared during their execution to identify the most suitable ones to solve the problem. This study concerns the assessment of the DMs' behavior during their execution because in general, works in the literature compare them only based on the quality of the obtained solutions.