The Pelican Optimization Algorithm (POA) is a nature-inspired met heuristic that has shown promise in solving complex optimization problems. This paper introduces an improved variant of POA, enhanced through the integration of Simulated Annealing (SA) and parallelized fitness evaluation, to boost its convergence efficiency and solution quality. The hybrid POASA algorithm leverages SA’s local search capabilities to refine candidate solutions during exploration efficiency. The performance of the improved POA is benchmarked against 23 standard test functions and compared with the original POA. The results demonstrate that the hybrid approach achieves greater optimization, reflecting improved convergence, solution stability, and robustness.