In computational mechanics of complex dynamical systems, robust design consists in finding designs of mechanical systems by solving nonlinear constrained optimization problems using numerical models which are little sensitive to uncertainties in the vicinity of the design point. All the published works in robust design concern data uncertainties and not model uncertainties. In the present work, a probabilistic methodology is proposed to solve the robust design optimization problem not only with respect to data uncertainties but also with respect to model uncertainties in the context of dynamical systems. The possible designs are represented by a numerical finite element model whose parameters belong to an admissible set of design variables. The nonparametric model of random uncertainties is used for taking into account model uncertainties and data uncertainties.
