Simulation-assisted response surface methodology

Simulation-assisted response surface methodology (SA-RSM) combines computer simulation models — such as finite element analysis, computational fluid dynamics, or discrete-event simulation — with the statistical framework of response surface methodology to efficiently map, model, and optimize system responses. Instead of running physical experiments, the researcher executes simulation runs at design points prescribed by an RSM design, fits a polynomial metamodel (surrogate) to the simulation outputs, and uses that metamodel to locate optimal factor settings.