Over the past few decades, modeling, simulation, and optimization tools have received attention for their ability to represent and improve complex systems. The use of metamodeling techniques in optimization via simulation problems has grown considerably in recent years to promote more robust and agile decision-making, determining the best scenario in the solution space. The objective of this article is to conduct a systematic literature review of metamodeling-based simulation optimization (MBSO). The main contributions of this paper are related to systematically gather, analyze, and discuss the knowledge disseminated in this area, support future research and expand the literature related to MBSO techniques. Research questions were planned to assist MBSO researchers and practitioners, by presenting the most frequent contexts, applications, methods, tools, and metrics found in metamodeling studies in optimization via simulation problems. We considered papers published in scientific journals and listed in the Web of Science, Scopus, ACM Digital Library, IEEE Xplore, and Science Direct databases. The conclusion related the gaps, opportunities, and future perspectives found during the development of this research, suggesting that this research area is growing in the past 15 years.

在过去的几十年中，建模、仿真和优化工具因其表示和改进复杂系统的能力而受到关注。近年来，通过模拟问题在优化中使用元建模技术已大大增加，以促进更稳健和敏捷的决策制定，确定解决方案空间中的最佳方案。本文的目的是对基于元建模的仿真优化 (MBSO) 进行系统的文献综述。本文的主要贡献是系统地收集、分析和讨论该领域传播的知识，支持未来的研究并扩展与 MBSO 技术相关的文献。研究问题旨在通过介绍最常见的背景、应用、方法、在通过模拟问题进行优化的元建模研究中发现的工具和指标。我们考虑了发表在科学期刊上并在 Web of Science、Scopus、ACM 数字图书馆、IEEE Xplore 和 Science Direct 数据库中列出的论文。结论与本研究发展过程中发现的差距、机遇和未来前景相关，表明该研究领域在过去 15 年中不断发展。