Data Envelopment Analysis Model with Decision Makers’ Preferences: A Robust Credibility Approach

Hashem Omrani*, Arash Alizadeh, Ali Emrouznejad, Tamara Teplova

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Data envelopment analysis (DEA) is one of the widely used methods to measure the efficiency scores of decision making units (DMUs). Conventional DEA is unable to consider both uncertainty in data and decision makers’ (DMs) judgments in the evaluations. This study, to address the shortcomings of the conventional DEA, proposes a new best worst method (BWM)- robust credibility DEA (BWM-RCDEA) model to estimate the efficiency scores of DMUs considering DMs’ preferences and uncertain data, simultaneously. First, to handle uncertainty in input and output variables, fuzzy credibility model has been applied. Additionally, uncertainty in constructing fuzzy sets is modeled using robust optimization with fuzzy perturbation degree. In this paper, two new types of RCDEA models are proposed: RCDEA model with exact perturbation in fuzzy inputs and outputs and RCDEA model with fuzzy perturbation in fuzzy inputs and outputs. In addition, to deal with flexibility of weights and incorporating DMs’ judgement into the RCDEA model, a bi-objective BWM-RCDEA model is introduced. Finally, the proposed bi-objective model is solved using min–max approach. To illustrate the usefulness and capability of the proposed model, efficiency scores of 39 distribution companies in Iran is investigated and results are analyzed and discussed. Finally, based on the results, recommendations have been made for policy makers.
Original languageEnglish
JournalAnnals of Operations Research
Early online date25 Oct 2021
Publication statusE-pub ahead of print - 25 Oct 2021

Bibliographical note

© Springer Nature B.V. 2021. The final publication is available at Springer via


Dive into the research topics of 'Data Envelopment Analysis Model with Decision Makers’ Preferences: A Robust Credibility Approach'. Together they form a unique fingerprint.

Cite this