Selecting the “best” using data envelopment analysis

One of the most important strengths of data envelopment analysis, (DEA), is that it allows almost complete freedom in the way that each decision making unit, (DMU), evaluates itself relative to its peers. This tends to result in many DMUs receiving a high efficiency score. Particularly when DEA is applied in a decision making context, it may be desirable to select a single option rather than determining the set of efficient alternatives in ranking efficient DMU or to assist selecting a “best” DMU. Several extensions to DEA have been proposed and used. This paper examines, compares and integrates a variety of these methods. A less complicated application area is used to investigate the subtleties of DEA cross-efficiency