A modified SBM-NDEA approach for the efficiency measurement in bank branches

Abstract

Efficiency measurement has been an essential topic in banking research. Many studies have used the data envelopment analysis approach as an effective tool for the efficiency measurement of bank branches. Recently, the Network (NDEA) method has been introduced, which involves internal processes and intermediate factors in the efficiency measurement. A few studies have used this method for the efficiency measurement of bank branches; they have considered two processes in branches and used two-stage models. In this study, we will use a more general approach and a more complete network structure consisting of three processes, including the deposit attraction process, deposit allocation process and banking services provision process. Since we want to obtain efficient targets, an envelopment form of the NDEA model had to be used. Therefore, a slack-based NDEA model, as introduced by Tone and Tsutsui (SBM-NDEA), was nominated to support its mathematical model. But according to the new categorization of the efficiency measurement factors introduced in this paper, and also regarding some previous reviews on SBM-NDEA model, the model will be modified to include the desired properties. Finally, by applying the modified model, branches efficiency scores and also efficient targets will be obtained.

A review of network DEA models based on slacks‐based measure: Evolution of literature, applications, and further research direction

“…As shown by the literature review, improvements and additions to the NDEA SBM were made that include several aspects that are observed in real-world processes, such as undesirable outputs, shared inputs, and the dynamic nature of the processes. Thus, Lin and Chiu (2013) proposed the integration of the NSBM and independent component analysis; Tone and Tsutsui (2014) with the dynamic network approach; Lozano (2015) and Boloori and Pourmahmoud (2016), considered the efficiency at the system level rather than at the process level as in Tone and Tsutsui (2009) On the other hand, it was possible to verify the studies that became the basis for the development of SBM-NDEA models and applications. Thus, from the analysis of the frequency with which each SBM-NDEA research is cited by others, which can be visualized through the number of connections presented in Fig.…”

Operational Research and Artificial Intelligence Methods in Banking

“…Moreover, a group of research focuses on improving the quality of bank efficiency estimations, using new advanced DEA-based approaches, such as two-stage models, slack-based and network approaches, free disposal hull models, non-convex models, and Bayesian approaches ( Holod & Lewis, 2011 ;Boloori & Pourmahmoud, 2016 ;Hasannasab et al, 2019 ;Tavakoli & Mostafaee, 2019 ;Tsionas, 2020 ). Moreover, the introduction of machine learning models has also heralded a new dawn for bank efficiency evaluation.…”

Performance, Efficiency, and Target Setting for Bank Branches: Time Series With Automated Machine Learning

“…SBM-NDEA is a linear programming technique, an advanced type of linear programming analysis, and it considers the interconnections of inputs and outputs with all variables. By applying the modified model, branch efficiency scores and efficient targets can be obtained via an innovative perspective [14]. In addition to systematic approaches [15], algorithmic and mathematical methods are also encountered in the literature.…”