Applying an enhanced hybrid interval slacks-based model to solve inverse DEA to estimate outputs

Abstract This paper focuses on inverse data envelopment analysis (DEA) under the following question:” If the inputs of $$\textit{DMU}_o$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mi>DMU</mml:mi> <mml:mi>o</mml:mi> </mml:msub> </mml:math> increase, how much should the outputs of $$\textit{DMU}_o$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mi>DMU</mml:mi> <mml:mi>o</mml:mi> </mml:msub> </mml:math> increase to decrease the efficiency score of $$\textit{DMU}_o$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mi>DMU</mml:mi> <mml:mi>o</mml:mi> </mml:msub> </mml:math> ? In these problems, a non-radial enhanced slack-based model is employed in the presence of hybrid interval data, which includes both continuous and integer interval data. Under this scenario, novel necessary and sufficient conditions to estimate outputs are considered based on the Pareto solution of related multi-objective nonlinear programming (MONLP). Furthermore, an example is presented to illustrate the problems associated with hybrid interval data. Finally, the conclusion and future work are presented