Comparison of Kriging-based algorithms for simulation optimization with heterogeneous noise

Many discrete simulation optimization techniques are unsuitable when the number of feasible solutions is large, or when the simulations are time-consuming. For problems with low dimensionality, Kriging-based algorithms may be used: in recent years, several algorithms have been proposed which extend the traditional Kriging-based methods (assuming deterministic outputs) to problems with noise. Our objective in this paper is to compare the relative performance of a number of these algorithms on a set of well-known test functions, assuming different patterns of heterogeneous noise. The conclusions and insights obtained may serve as a useful guideline for researchers aiming to apply Kriging-based methods to solve engineering and/or business problems, and may be useful in the development of future Kriging-based algorithms.