The increasing complexity of optimization problems and the rapid increase in data volumes have made traditional sequential computation ineffective in providing quick solutions. The present work focuses on the potential of parallel computing architectures to solve large-scale optimization problems. There has been significant progress in hybrid metaheuristic frameworks that combine evolutionary algorithms with local search approaches. Parallel techniques optimize computing resources better while balancing exploration and exploitation. Surrogate modeling has proven effective in lowering the computational cost of expensive objective function evaluations. The deployment of parallel computing infrastructure and distributed computing environments has yielded measurable performance gains. However, significant obstacles remain. Communication overhead between processes limits speed, and load balancing between processors is still an issue. Decomposing strongly coupled optimization problems into parallelizable components is difficult. This study synthesizes current advances and identifies key research directions in the convergence of optimization theory, high-performance computing, and simulation-based modeling.