Network fuzzy data envelopment analysis

Abstract

In this chapter a general approach to handle fuzzy data when the units under analysis are formed by a network of processes is presented. Conventional DEA assumes a single-process that consumes all the different inputs and produces all the different outputs. Network DEA, on the contrary, considers different interrelated processes, each one with its own inputs, its own outputs and, very important, its own technology. This allows a more fine-grained analysis although at the expense of requiring more data. Conventional Network DEA approaches assume crisp data although recently two proposals have been made that can process fuzzy data in the special cases of a serial two-stage system and of parallel production processes. There is, however, a need to deal with general networks of processes which can have fuzzy input or output data. In this chapter, several Fuzzy DEA approaches are extended to Network DEA. The resulting models are illustrated on a dataset from the literature. © 2014 Springer-Verlag Berlin Heidelberg.