Article ID: | iaor20173157 |
Volume: | 255 |
Issue: | 1 |
Start Page Number: | 323 |
End Page Number: | 346 |
Publication Date: | Aug 2017 |
Journal: | Annals of Operations Research |
Authors: | Cook Wade, Zhu Joe, Du Juan |
Keywords: | statistics: data envelopment analysis, ecology, performance |
Electricity generation currently is the main industrial source of air emissions in the United States. Both researchers and practitioners are interested in conducting studies to evaluate the ecological performance of this industry, in order to propose solutions to curb emissions of air pollutants and to improve the efficiency of converting fossil resources into electric energy. In this paper, data envelopment analysis (DEA) is used to assess ecological efficiency where air emissions are used as undesirable outputs. Although conventional DEA does not require a priori information on the input and output weights, weight restrictions can be incorporated to reflect a user’s preference over the performance metrics, or to refine the DEA results. Adding weight restrictions voids the fact that DEA scores are independent of the units of measurement. To incorporate weight constraints in ecological efficiency assessment, this paper develops a DEA model that is units‐invariant when weight restrictions are imposed. Moreover, the proposed model is equivalent to the standard units‐invariant DEA model when weight restrictions are not present.