Performance Evaluation of DMU's Using Window DEA and Decision Tree

Efficiency is an issue of importance and interest to both managers of different organizations and customers who use the products and services of these organizations. The aim of this research is to study the efficiency of pharmaceutical companies accepted in the Stock Exchange Organization by using Window Data Envelopment Analysis (WDEA) and then, to provide some rules based on the decision tree. Window Data Envelopment analysis method, partly resolves the problem of inadequacy of the observations by enabling combination of time-series and cross-sectional observations. This method acts on the basis of moving average and it is of use for finding performance trends of a unit over time. In this research, regarding the inputs and outputs and using the method of Window Data Envelopment Analysis, efficiency of 23 pharmaceutical companies accepted in Stock Exchange Organization was evaluated in the state of constant returns to scale during 2009-2014, and then the results obtained were used as the class label of Decision Making Units (DMU’s), which are in fact inputs of decision tree metho d. Finally, using the decision tree, rules implicit in the data, are extracted.