A new methodology for sustainability measurement of banks based on rough set theory

Abstract

Sustainability measurement of banks is a complicated process, mainly dependent on the intuition of banking and sustainability experts. In this paper, a Decision Support System (DSS) is designed, which can accurately determine the corporate sustainability level of banks based on the banks’ self-declaration of sustainability indexes. The goal of this DSS is to predict the sustainability level of a bank by using self-declaration data provided by top managers of a bank before starting the costly and time-consuming process of sustainability auditing to help the assessment team for better planning before starting the process. In order to create such a system, sustainability auditing processes of 8 banks were investigated, and subsequently, 29 condition attributes and one decision attribute are determined as decision system of the rough set. ROSETTA software is used to extract 16 rule models regarding different discretization methods, reduct generating techniques, and rule generating approaches. The rules that are generated from the Naive/Genetic/object-related model (with the considerably high accuracy of 91.5%) are selected as the inference engine of DSS.