In this paper, an integrated algorithm is presented to assess the efficiency of electric companies using data envelopment analysis (DEA) combined with fuzzy c-means clustering (FCM) and principle component analysis (PCA). FCM algorithm is applied to find similar distribution companies and cluster similar companies into different groups. One of the important steps in the design of the benchmarking model is selecting input and output variables. PCA is used to reduce the number of input and output variables under study. DEA efficiency is highly sensitive to errors in the data, so a bootstrap method is used to assess this uncertainty by estimating bias and confidence intervals. The results of benchmarking include DEA efficiency scores (overall, technical, and scale efficiency), bootstrap technique, slacks in inputs and outputs of inefficient companies, and sensitivity analysis.
展开▼
