Based on the proposed method of Tong et al. (2008), this paper provides a robust design procedure to identify optimal factor levels of multiple quality responses for dynamic systems. The reciprocal coefficient of variation is used to compute the relative efficiency of decision making unit via the constant return to scale data envelopment analysis model. The proposed method releases the limitations of Taguchi method and allows a nonlinear function to model the relationship between the multiple quality responses and signal factors to reach a robust design. The application of the proposed methodology is illustrated through a real example.