Data Envelopment Analysis

José H. Dulá

ABSTRACT

Data Envelopment Analysis (DEA) is a nonparametric frontier estimation methodology based on linear programming for measuring relative efficiencies of a collection of firms or entities in transforming their inputs into outputs. DEA is a standard tool in OR/MS whose number and domain of applications is rapidly growing. This chapter introduces DEA, focusing on the algorithmic and computational aspects of the methodology. As more analysts apply the methodology and as studies grow in scale and sophistication, data sets become larger, analyses become more frequent, even dynamic, and more effort is spent on extracting information efficiently. These issues place algorithmic and computational aspects of DEA at the forefront.