Allocating fixed resources for DMUs with interval data

Abstract

Conventional DEA models tend to allocate the fixed resources to multiple decision-making units (DMUs) and treat the allocated resource as an extra input for every single DMU. However, the existing DEA resource allocation (DEA-RA) methods are applicable exclusively to the DMUs with exact values of inputs and outputs. A lack of precision for the input or output data of DMUs, such as the interval data, would cause a failure of the existing methods to allocate resources to DMUs. In order to resolve this problem, three DEA-RA models are proposed in this paper for different scenarios of decision-making. All of the proposed DEA-RA models are based on a set of common weights. Finally, the proposed models are empirically tested and validated through three examples. As revealed by the results, our proposed models are capable of providing a more fair and practical initial allocation scheme for decision makers.