Fuzzy clustering of homogeneous decision making units with common weights in data envelopment analysis

Sajad Kazemi, Reza Kiani Mavi, Ali Emrouznejad, Neda Kiani Mavi

Research output: Contribution to journalArticlepeer-review


Data Envelopment Analysis (DEA) is the most popular mathematical approach to assess efficiency of decision-making units (DMUs). In complex organizations, DMUs face a heterogeneous condition regarding environmental factors which affect their efficiencies. When there are a large number of objects, non-homogeneity of DMUs significantly influences their efficiency scores that leads to unfair ranking of DMUs. The aim of this study is to deal with non-homogeneous DMUs by implementing a clustering technique for further efficiency analysis. This paper proposes a common set of weights (CSW) model with ideal point method to develop an identical weight vector for all DMUs. This study proposes a framework to measuring efficiency of complex organizations, such as banks, that have several operational styles or various objectives. The proposed framework helps managers and decision makers (1) to identify environmental components influencing the efficiency of DMUs, (2) to use a fuzzy equivalence relation approach proposed here to cluster the DMUs to homogenized groups, (3) to produce a common set of weights (CSWs) for all DMUs with the model developed here that considers fuzzy data within each cluster, and finally (4) to calculate the efficiency score and overall ranking of DMUs within each cluster.
Original languageEnglish
Pages (from-to)813-832
Number of pages20
JournalJournal of Intelligent and Fuzzy Systems
Issue number1
Early online date23 Oct 2020
Publication statusPublished - 2021

Bibliographical note

Copyright ©2020 The Authors. Kazemi, Sajad et al. ‘Fuzzy Clustering of Homogeneous Decision Making Units with Common Weights in Data Envelopment Analysis’. 1 Jan. 2020 : 1 – 20. DOI: 10.3233/JIFS-200962


  • Clustering
  • Common set of weights (CSW)
  • Data envelopment analysis
  • Fuzzy DEA
  • Non-homogeneous


Dive into the research topics of 'Fuzzy clustering of homogeneous decision making units with common weights in data envelopment analysis'. Together they form a unique fingerprint.

Cite this