A Discriminant Analysis and Goal Programming Approach to solve the Multiple Criteria Data Envelopment Analysis Model

Data Envelopment Analysis (DEA) is a method measuring the relative performance of a group of decision making units (DMUs) which consume a number of inputs and produce several outputs at different quantities. In spite of its popularity, DEA still endures some kinds of shortcomings. For instance, DEA lacks the discriminating power among efficient DMUs. In this paper, we introduce a method which utilizes goal programming and discriminant analysis to solve the multiple criteria DEA model. The proposed method develops a classification function which separates efficient and inefficient DMUs and generates an efficiency ranking for all DMUs. Furthermore, it allows decision-makers to incorporate a priori information about the factor weights via proportional virtual weights restrictions or other forms of weights restrictions. Performance of the proposed method is illustrated by two real applications, which have been studied in the literature.