Soccer game optimization with substitute players

• Player substitution mechanism is proposed to enhance the Soccer Games Optimization.
• The proposed method is compared to SGO without substitution.
• The experiment results confirm the proposed method is better.
• We investigate the method sensitivity for unimodal and multimodal functions.

Metaheuristic methods are powerful optimization methods to solve various problems and have been used widely. In this paper, a player substitution mechanism to enhance the performance of the Soccer Game Optimization (SGO) is proposed. The substitution mechanism refers to the competition concept existing in the evolutionary algorithm. The substitute players are continuously updated to keep up with a set of best solutions so far. This paper discusses the sensitivity of its parameters as well as assesses the performance of the proposed method using unconstraint continuous problems. The proposed method is compared to the SGO without player substitution, Biogeography-based optimization (BBO), Ant Colony Optimization (ACO), Differential Evolution (DE), Evolutionary Strategies (ES), Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The experiment results show that the proposed method performs better than other algorithms used in this paper.