Kriging, the fitting of a Guassian process model to data, has becoming widely used for creating a metamodel for complex simulation models. However, it has been observed that various software packages give somewhat different results when fitting the same Gaussian process model to the same data. The goal of this study is study and compare the different software options for kriging as a metamodel for simulation data. Some alternative methods to kriging will also be identified and compared to the kriging models. The purpose of the study is to provide guidance to our future research in the design and analysis of experiments for fitting metamodels to simulation data.
|Effective start/end date||9/29/15 → 9/28/16|
- Naval Supply Systems Command Fleet Logistics Center San Diego (N00244-15-2-0004)