Network DEA models in the presence of integer-valued data and flexible factors: a slacks-based approach
摘要
Traditional data envelopment analysis (DEA) models are concerned with measuring the relative efficiency of units with respect to multiple inputs and multiple outputs. These models assume real-valued inputs and outputs. In these models, the role of the measures in relation to the input and/or output is known, and these models also neglect the intervening or linking activities. However, in the real world, there are situations where some inputs and/or outputs can only take integer values, and a measure can also play the role of either an input or an output. In this paper, we propose network DEA models that can handle intermediate products when some inputs and/or outputs have integer values and flexible factors are available.