Data Envelopment Analysis (DEA)

Abstract

Measurement of efficiency of any organisation (e.g., hospital, bank, etc.) that uses multiple inputs and generates multiple outputs is complex and comparisons across units are difficult. Charnes and Cooper (1985) describe a non-parametric approach in such situations to measure efficiency and the technique is known as data envelopment analysis (DEA). This analysis method is basically a linear pro gramming-based technique used for measuring the relative performance of organisational units where the presence of multiple inputs and outputs makes comparisons difficult. It involves identification of units, which in relative sense use the inputs for the given outputs in the most optimal manner; DEA uses this information to construct efficiency frontiers over the data of available organisa tion units. This efficient frontier is used to calculate the efficiencies of the other organisation units that do not fall on the efficient frontier and provide informa tion on which units are not using inputs efficiently. The objective of this article is to introduce the technique and demonstrate it through an example to show how relative efficiencies can be determined and identify units that are relatively less efficient.