centralized resource allocation and reduction environmental pollution using dea models

While conventional models of data envelopment analysis (dea) set targets for each dmu separately, there are many models based on dea, which resource allocation to a set of dmu. these dmus organization with a centralized decision maker. some researchers presented resource allocation models as a multiple-objective linear programming (molp), so that paying more attention to energy saving and environmental pollution reduction. in order to allocate resources reasonably, researchers consider boundaries for desirable and undesirable outputs changes, which are selected by dm. in this paper we show that if these boundaries are not suitable, the model will be infeasible. also, the model is modified by goal programming in such a way that it becomes feasible. moreover, since the undesirable output is considered as a function of total desirable outputs, so if necessary, the changes should be done on total desirable outputs. the result of this model is that the reduction proportion of total environmental pollution emissions for saving energy is bigger than the total of desirable outputs reduction proportion.