LAUSR

Abstract Data envelopment analysis (DEA) is a well-known and widely used method for performance evaluation in a set of homogeneous units. We propose a new bi-level DEA model for efficiency measurement and target setting. The fundamental novelty of the proposed model is threefold. We: (1) set both efficiency and profit concurrently as targets; (2) limit the amount of changes in the inputs and outputs to prevent unachievable targets; and (3) predict some targets for efficient units beyond the inefficient ones. We present a case study in the banking industry to demonstrate the efficacy of efficiency measurement and target setting in the proposed models.