One of the most prevalent problems in linear programming as one of the convenient models in the field of operation research environment is Data Envelopment Analysis (DEA), which supports the efficiency of Decision-Making Units (DMUs). Usually, accurate data are common; however, in the real world, we are facing an inaccurate situation. In this paper, a new model for assessing DMUs in a fuzzy environment is presented; we consider the inverse DEA model with the variable return to scale with fuzzy numbers for fluctuating data. A case study is given to illustrate its performance.
KEYWORDSfuzzy mathematical programming; fuzzy ordering; data envelopment analysis 1 Introduction D ata Envelopment Analysis (DEA) is a technique for deliberating the performance of Decision-Making Units (DMUs) with numerous inputs and outputs. Each DMU uses several inputs to generate several outputs. Moreover, the performance of DMUs is evaluated based on obtained inputs and outputs. DEA has been widely used in many fields of science. Pascoe et al. [1] used DEA to assess management alternatives in the presence of multiple objectives. Examples of the use of DEA to assess the financial effectiveness of insurance companies are presented in Ref. [2]. Nasseri and Khatir [3] organized a two-stage DEA model by taking into account undesirable output with fuzzy stochastic data. As one step forward, the Inverse of the DEA (IDEA) model has been recently introduced by Wei et al. [4] so that it tries to answer the question: If input (output) in a DMU is agitated, then how output (input) should be changed to keep the relative performance of the DMU? IDEA models have different applications in practical cases [5−7] . Furthermore, the IDEA model with a variable return to scale (Inverse Banker, Charnes, and Cooper model (IBCC)) is suggested by Lertworasirikul et al. [8] Next, Ghiyasi [9] proposed several problematic issues, and he challenged that the suggested Multi-Objective Linear Programming (MOLP) model by Lertworasirikul et al. [8] is not as effective as it was claimed.Fuzzy uncertainty, grey uncertainty, and rough uncertainty are intertwined facets of imprecision in