Metamodeling of combined discrete/continuous responses

Metamodels are effective for providing fast-running surrogate approximation s of product or system performance. Because these approximations are genera lly based on continuous functions, they can provide poor fits of discontinu ous response functions. Many engineering models produce functions that are only piecewise continuous, due to changes in modes of behavior or other sta te variables. The use of a state-selecting metamodeling approach that provi des an accurate approximation for piecewise continuous responses is investi gated. The proposed approach is applied to a desk lamp performance model. T hree types of metamodels, quadratic polynomials, spatial correlation (krigi ng) models, and radial basis functions, and five types of experimental desi gns, full factorial designs, D-best Latin hypercube designs, fractional Lat in hypercubes, Hammersley sampling sequences, and uniform designs, are comp ared based on three error metrics computed over the design space. The state -selecting metamodeling approach outperforms a combined metamodeling approa ch in this example, and radial basis functions perform well for metamodel c onstruction.