How to deal with numbers of decision making units and variables in data envelopment analysis

Abstract

Sufficient numbers of Decision Making Units (DMUs) in comparison with the number of input and output variables has been a concern of using Data Envelopment Analysis (DEA) in the last three decades. There are several studies in the literature of DEA which have tried to handle this issue by providing additional procedures to increase the number of DMUs, decreasing the number of variables or finding a relationship between the number of DMUs and variables. However, there are no concerns about the number of DMUs in comparison with the number of variables while Kourosh and Arash Method (KAM) is applied. A geometric reason is provided to depict the validity of the method without any extra conditions or additional methodologies. The technique is quite simple with no computational complexities of current methodologies even if the number of DMUs is less than the number of variables. A real-life numerical example of 32 DMUs with 45 variables demonstrates the advantages of the proposed technique.