Application of fuzzy network data envelopment analysis model with optional input-undesirable output in order to evaluate the performance of bank branches

The issue of efficiency evaluation is one of the most important challenges facing managers in the dynamic and vital banking industry. Due to the very important and fundamental role of banks in most economic activities, the study of bank efficiency has a special place. Models based on data envelopment analysis had wide applications in the field of measuring and evaluating the efficiency of banks. In this study, we tried to add other assumptions to the traditional model of data envelopment analysis, using a model that is more consistent with the real conditions related to decision-making units and calculates the efficiency more accurately. In designing the data envelopment analysis model in the present study, attention has been paid to the networking and internal relations of each decision-making unit, undesirable output, involuntary input and the fuzzy nature of the variables. After developing the model with the mentioned assumptions, first, based on reviewing previous studies as well as field studies and obtaining opinions from industry experts, indicators considered as input and output. After identifying the indicators, fuzzy Delphi method used for initial screening of the identified indicators from the research literature. After the initial approval of the indices, the indices analyzed using the confirmatory factor analysis technique. Finally, the research model solved using data collected by GAMS software and with alpha-cutting approach. The results show that out of 38 branches examined, 8 branches are efficient and another 30 branches are inefficient.