An enhanced interval slacks-based model to solve inverse DEA for estimating outputs

Abstract

This study proposes a non-radial slacks-based Inverse Data Envelopment Analysis (Inverse DEA) framework for decision-making units under interval data. To address uncertainty in both inputs and outputs, the methodology integrates interval arithmetic with partial ordering, enabling robust efficiency evaluation. The research investigates how output levels should be adjusted when inputs increase, and the inefficiency score of a unit decreases by a specified proportion. An enhanced inverse DEA model is developed, and necessary and sufficient conditions for output estimation are derived through the Pareto optimal solutions of a multi-objective nonlinear program. Two illustrative examples are provided to illustrate the applicability of the proposed approach under interval data scenarios.