Multivariate modelling of complex simulation-based systems

Complex computer simulations can take a long time to complete. Often one can vary several input parameters for the simulation. In such case, one wants to know the output of the simulation as a function of the input parameters in order to characterize and explore the whole design space. Traditionally, simulations are performed on a rectangular grid in the input space. This approach quickly becomes infeasible as too many simulations are needed. This paper presents an adaptive methodology to capture the complex input-output behaviour of the simulator in a multivariate surrogate model.