Two-stage Serial Data Envelopment Analysis Models: Comparison of Approaches

Data envelopment analysis (DEA) is a modelling tool for assessment relative efficiency and performance of the set of homogeneous decision making units (DMUs) that transform multiple inputs into multiple outputs.Traditional models consider one-stage transformation -DMUs are black boxes that use multiple outputs and produce multiple inputs.In the contrary, network DEA models assume production process in a more general and complex way.In two-stage serial DEA models, the production process consists of two stages.The inputs of the first stage are used for production of the first stage outputs.These outputs enter the second stage as inputs and are used for production of the final outputs of the production process.The aim of this paper is to compare the most important approaches for evaluation of efficiency of the two-stage serial production processes based on the methodology of DEA.The properties of the models are discussed.A numerical example illustrates the results of all models.