Centralized resource allocation DEA models based on revenue efficiency under limited information

In this paper, we extend the centralized DEA models by Lozano et al (2011) to allocate resources based on revenue efficiency across a set of DMUs under a centralized decision-making environment. The aim is to allocate resources so as to maximize the total output revenue produced by all the DMUs under limited information. To uncover the sources of total revenue increase from the centralized resource allocation model, we further decompose the aggregate revenue efficiency into three components: the aggregate output-oriented technical efficiency, the aggregate output allocative efficiency and the aggregate revenue re-allocative efficiency. Finally, two empirical data sets are presented to show that our proposed approach is not only an efficient tool to allocate the resources among the DMUs based on the revenue efficiency but additionally provides the central DM with guidance on how to identify the weak areas where more effort should be devoted to improve the total outputs.