For an optimization problem, when the evaluation function costs in term of resource, chaining the construction of a surrogate model with the optimization process could be a good solution. Uranie provides modules that can be used directly to do so. The Metamodel Optimization module gives another way to do similar things by coupling the two process: the surrogate model construction and the optimization research.

So far, the module provides a parallelized version of the Efficient Global Optimization (EGO) algorithm for mono objective problems.