Adaptive metamodeling simulation optimization: Insights, challenges, and perspectives

A pillar of Industry 4.0, Simulation Optimization is a powerful tool used across several fields, enabling system evaluation under varying conditions, facilitating performance analysis, and more efficient decision-making. On the other hand, the simulations might be time-consuming, particularly when considering complex model optimization. In this sense, metamodeling has emerged as a promising technique for simulation optimization. Metamodeling aims to establish and estimate a relationship between the inputs and outputs of a simulation model, creating a simplified model used to evaluate potential solutions during the optimization process. Meta- modeling approaches can be classified as metamodeling with a fixed experimental design and adaptive meta- modeling. This paper presents a systematic literature review of adaptive metamodeling in simulation optimization problems. The primary contributions of this paper are the systematic collection, examination, and discussion of knowledge disseminated in this field. We aim to support upcoming research endeavors and enhance the existing literature concerning adaptive metamodeling techniques. Our scope encompassed scientific journal papers cataloged in Scopus, Web of Science, IEEE Xplore, ACM Library, Taylor & Francis, and Science Direct databases. The research questions intend to aid researchers and practitioners in summarizing prevalent contexts and methods addressed in adaptive metamodeling studies within simulation optimization problems. As a result, we discuss the main metamodels algorithms, optimization and simulation methods, acquisition functions, and sampling techniques. This paper also provides guidelines for adaptive metamodeling studies by highlighting the fundamental, suggested, and optional steps to be followed in new adaptive metamodeling applications. In the conclusion, we addressed the identified gaps, opportunities, and prospective directions about the theme.