Performance Measurement of Decision Making Units with Network Structure in the Presence of Undesirable Output

In the performance evaluation process, using the classic data envelopment analysis (DEA) models, decision making units (DMUs) are considered as black boxes. While in many cases and different applications such as investment funds, banks, insurance companies, etc., DMUs have a network structure. In addition, in many network structures, some of the indicators used to calculate the efficiency of DMUs have an undesirable nature. Therefore, it is necessary to provide an effective approach that takes into account the internal structure of the DMus and the inter-organizational relationships in the calculation of efficiency and also is capable to be used in the presence of undesirable data. Accordingly, the purpose of the present study is to provide a network data envelopment analysis (NDEA) model in order to performance assessment of DMUs with a two-stage structure in the presence of undesirable outputs. It should be noted that according to the two-stage structure of investment companies and the presence of undesirable output in their evaluation process, ten investment companies from Tehran Stock Exchange have been selected as a case study for the research. Finally, the results of their performance evaluation by applying the proposed approach of research have been analyzed.