Complex simulation models can be time-consuming and resource-intensive, especially when dealing with uncertainty. To tackle this issue, researchers compared three common metamodels (mathematical models) and different sampling methods to optimize performance. They tested these methods on various problems, considering both small and large sample sizes. The goal was to find the best combination of metamodels and sampling techniques for robust optimization under uncertainty. The study found that the performance of metamodels in robust optimization is influenced by sample size, and provided guidelines for practitioners to choose the most effective approach for finding optimal solutions in simulation-optimization problems.