A novel approach for solving multi-objective optimization problems is proposed. The algorithm is able to compute non-convex as well as convex Pareto fronts of large-scale dynamical optimization problems. Problems induced by inequality constraints are handled. The algorithm is tested successfully on two case studies.