This study presents the DFSS optimization for the dynamic responses of a paper feeding mechanism. The flexible paper is idealized as a series of rigid bars connected by revolute joints and rotational spring dampers. In this mechanism, a paper is fed by a contact and friction mechanism on rollers or guides. The design objective is to minimize the slip amounts between paper and mechanisms and satisfy the 6-sigma constraint for the nip forces of rollers. In order to avoid the difficulty of design sensitivity analysis and overcome the numerical noise, a meta-model based optimization is employed. In this approach, first, the space filling methods and the classical DOE methods are used to generate sampling points. Second, the meta-models are constructed from the Kriging, RBF and RSM methods. Finally, a well-developed numerical optimizer sequentially solves the approximate optimization problem. In the numerical test, the DFSS for the paper feeding mechanism problem, having 8-random design variables, is solved in only 23 analyses.