Robust efficiency in data envelopment analysis with VRS technology

Abstract

One of the fundamental problems in the classic DEA is lack of ability to distinguish unit's performance scores that is considered as a disadvantage. Recently, Parkan et al. [9] tried to address this problem. They proposed to assess each unit both optimistic and pessimistic views are taken into account. In contrast to traditional evaluation, one index is considered for each unit based on the lowest measured performance that is called robust efficiency. In this way, a new technology was made on the assumption of constant returns to scale. In this paper, the production technology with variable returns to scale assumptions made and the corresponding models are formulated as linear programming. Finally, it is shown that, the models based on the robust efficiency has more discriminatory power of the classic DEA.