This project focuses on developing methods to enhance the scalability and sampling efficiency of metamodel-based simulation analysis.  The key technical components include (1) a model-free method based on the Morris elementary effects method for sequential input screening with rigorous statistical performance guarantees,  (2) theory and methods for constructing knowledge- and data-driven scalable heteroscedastic metamodels, and (3) a metamodel-based global sensitivity analysis approach to quantifying the impact of each active input under heteroscedasticity. 

This project is supported by the National Science Foundation CAREER under Grant No. CMMI-1846663.