This paper presents a novel metaheuristic algorithm aimed at addressing the inefficiencies of conventional approaches that use a single operator and unique population to handle both exploration and exploitation. To achieve a more effective search process, the algorithm introduces a structured population that is divided into two distinct groups: explorers ( \(XPL\) ) and exploiters ( \(EXP\) ). \(XPL\) agents perform global exploration by following complex trajectories generated from randomly selected Bézier functions, promoting diverse and comprehensive coverage of the search space. In contrast, the \(EXP\) agents are responsible for local refinement, updating their positions through a specialized hexagonal crossover operator. This operator generates new candidate solutions within a hexagon-shaped region by incorporating information from other high-quality individuals in the population. By integrating these two mechanisms, each dedicated to either exploration or exploitation, the algorithm enhances its ability to focus on promising regions, while maintaining global diversity. The proposed method was compared with several state-of-the-art metaheuristics using a suite of shifted functions specifically designed to challenge center-biased strategies. The experimental results demonstrate that the algorithm consistently produces a superior solution quality and faster convergence while effectively avoiding the typical flaws of center-biased methods.
