Environmental pollution has caused governments to be concerned about energy saving and the reduction of environmental pollution. Some researchers have presented resource allocation models as multi-objective linear programming (MOLP) in order to pay more attention to energy saving and environmental pollution reduction. Energy saving affects both desirable and undesirable outputs. In this paper, we argue for the inapplicability of the existing models for reducing the undesirable outputs through energy saving. The purpose of this paper is to design a model based on data envelopment analysis (DEA) that would result in reduced pollution through energy saving. Moreover, since an undesirable output is considered as a function of the total desirable outputs, if necessary, the changes should be applied to the total desirable outputs and there is no need to reduce each desirable output individually. Finally, the model proposed based on goal programming (GP) is used in 20 different regions in China. The results produced by this model indicate that the reduction proportion of total environmental pollution emissions per energy saving was larger than the reduction proportion of total desirable outputs.