A data envelopment analysis-based clustering approach under dynamic situations

Cluster analysis is one of the most useful tools for exploring the underlying structure of a given data set and is being applied in a wide variety of engineering and scientific disciplines. And Data Envelopment Analysis (DEA) is a data-driven tool for the efficiency evaluation of homogeneous decision-making units (DMUs). Few studies paid attention to the combination of the clustering approach and DEA. The paper proposes a method of clustering DMUs by the production function under dynamic situations. The paper determines the reference vector using the extreme efficient DMUs, defines the asymmetric dissimilarity, and realizes the clustering by a hierarchical algorithm. The numerical experiments are illustrated to examine the validity of the proposed method, and the experiments show that the method gives reasonable results. The proposed method is the first study for clustering time-varying DMUs by the production function.