Efficiency evaluation of electricity distribution companies: Integrating data envelopment analysis and machine learning for a holistic analysis

Evaluating the efficiency of electricity distribution companies (EDCs) accurately is one of the most important issues for regulators and policy makers. This research combines the results of data envelopment analysis (DEA) and corrected ordinary least squares (COLS) with machine learning techniques to evaluate a set of EDCs in the period 2011–2020. We propose a three-stage process. First, for each year, the efficiency scores of EDCs are measured using DEA and COLS methods. Then, this study applies support vector regression (SVR), a powerful machine learning technique, to estimate the efficient frontier and to calculate the efficiency of the EDCs. The efficiencies generated by DEA, COLS, and SVR are not the same and are used to construct fuzzy triangular numbers. Finally, the fuzzy efficiencies are considered as criteria for the technique for order performance by similarity to the ideal solution (TOPSIS), and the final efficiencies and ranks are obtained using the fuzzy TOPSIS (FTOPSIS) method. In addition, using the fuzzy C-means clustering (FCM) algorithm, the EDCs are clustered and discussed. The results show that there are increasing and decreasing trends for the selected EDCs in the period 2011–2022. In addition, some EDCs act in a poor situation and their performance should be improved.