The impact of decision-making units features on efficiency by integration of data envelopment analysis, artificial neural network, fuzzy C-means and analysis of variance

Abstract

In today's working environment, there is a great desire to identify the critical attributes for sensitivity analysis of inefficient decision-making units (DMUs) regarding personnel attributes. An integrated algorithm, which uses data envelopment analysis (DEA) and data mining tools including fuzzy C-means (FCM), rough set theory (RST), artificial neural network (ANN), cross validation test technique (CVTT) and analysis of variance (ANOVA), is proposed to asses the impact of personnel attributes on efficiency. DEA is used for DMUs' efficiency evaluation. ANN is employed with regard to its ability to model linear and non-linear systems. As numerous inputs are not useful for ANN modelling, RST and ANN are combined to resolve this issue. RST is used to decrease the time of decision-making. FCM is used for data clustering and finally ANOVA is utilised for identification of attributes importance. The proposed algorithm is applied to an actual banking system.