Optimization problems arise in all scientifical disciplines which work with unknown parameters, such as economical mathematics, statistics and engineering. For most problems -e.g. characterized by nonlinearity, multimodality etc.- evolutionary algorithms have proved to be promising optimization algorithms. Nevertheless, the performance of the global optimal solution located by optimization methods can be quite dependent on the given boundary conditions and consequently very sensitive to environmental uncertainties so that the off-design performance may deteriorate noticeably. Accounting for robustness against environmental uncertainties and the evolutionary optimization towards an operating range in opposite to an operating point is the focus of this book. The implicit averaging is supposed to be a highly efficient method which randomly varies the operational conditions over generations. The explicit averaging is a rather approved but more computational expensive procedure using additional fitness evaluations per generation. In this book both methods are applied to generated test functions as well as an aerodynamic optimization problem. The latter is combined with meta-modeling.