Multiplier bounds in DEA via strong complementary slackness condition solution

Abstract Data envelopment analysis (DEA) is a methodology for evaluating the relative efficiency of peer decision making units (DMUs) with multiple inputs and multiple outputs. In DEA evaluation, the efficiency scores for inefficient DMUs are obtained by calculating the proportional input or output Changes required to reach the DEA efficient frontier. It is likely that further individual input and output changes can be made to improve the performance. Such individual changes are called DEA slacks which also represent inefficiency. Methods have been developed to deal with the non-zero DEA slacks by re-shaping the original DEA frontier. This paper develops an alternative approach to eliminate the non-zero DEA slacks while keeping the original DEA frontier unchanged.