A new DEA model for slacks-based measure of efficiency and super-efficiency with strongly efficient projections

Most of the existing researches for the integrated model of slack-based measure (SBM) and SuperSBM models ignore two major problems. First, the SuperSBM model may yield weakly efficient projections such that the super-efficiency is overestimated due to ignoring the inefficiency components. Second, in the existing integrated models, some integrated models have feasible solutions, but their scale is very large; some integrated models are very small in scale, but there may be no feasible solutions. Therefore, this paper tackles these problems and presents a modified SuperSBM (MSuperSBM) model. The MSuperSBM model can differentiate all decision-making units and determine their strongly efficient projections simultaneously. And the actual-efficiency scores obtained objectively discount the inefficiency components relative to the strongly efficient point, thereby they are not overestimated. In addition, the MSuperSBM model contains only necessary decision variables and constraints, which can reduce the number of required computations for large-scale performance evaluation. Finally, through several numerical experiments, the advantages of our model in the computational results and computational scale are verified.

Automated summary

This study addresses two major problems in the existing research on the integrated model of slack-based measure (SBM) and SuperSBM models, and proposes a modified SuperSBM (MSuperSBM) model. The MSuperSBM model can differentiate decision-making units and determine their strongly efficient projections, thereby accurately evaluating efficiency scores without overestimation. Moreover, the model only includes necessary decision variables and constraints, reducing the computational complexity for large-scale performance evaluation.