Soft Computing https://doi.org/10.1007/s00500-018-3007-2 FOCUS Input–output performance efficiency measurement of an electricity distribution utility using super-efficiency data envelopment analysis Miriam F. Bongo 1 · Lanndon A. Ocampo 2 · Yannie Ann D. Magallano 3 · Geraldine A. Manaban 3 · Ezra Kim F. Ramos 3 © Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract This paper applies the conventional DEA model and super-efficiency analysis in measuring the efficiency of an electricity distribution utility which involves 12 power lines as DMUs. The input indicators considered are purchased electricity supply and the total length of power lines, while electricity consumed, the number of consumers, and total power losses are the output indicators. The results revealed that 4 out of the 12 power lines are inefficient and thus need to be improved. The model provided a guideline how these inefficiencies may be addressed by means of benchmarking. Keywords Efficiency · Data envelopment analysis · Super-efficiency DEA · Electricity distribution utility 1 Introduction Electricity plays a key role in keeping homes and businesses running smoothly (Wang et al. 2013). It is very important in terms of social equality and development (Yilmaz et al. 2015). Electricity supply is considered as a public service in many countries; thus, it has to be delivered with high-quality standards in an efficient and productive way. Communicated by A. Genovese, G. Bruno. B Miriam F. Bongo miriam.bongo@yahoo.com Lanndon A. Ocampo lanndonocampo@gmail.com Yannie Ann D. Magallano annyiedeveyra@gmail.com Geraldine A. Manaban manabangeraldine@gmail.com Ezra Kim F. Ramos ezrakimramos@gmail.com 1 Department of Mechanical and Manufacturing Engineering, University of San Carlos, 6000 Cebu City, Philippines 2 Department of Industrial Engineering, Cebu Technological University, Corner M.J. Ave. & R. Palma St., 6000 Cebu City, Philippines 3 Department of Industrial Engineering, University of San Carlos, 6000 Cebu City, Philippines An electricity distribution utility (EDU) consists of a very complex system. It involves four processes, namely genera- tion, transmission, distribution, and consumption that need to be considered in terms of overall efficiency performance (Tavassoli et al. 2015). Assessment in EDU is one of the most important issues among regulators, especially in electricity restructuring and reform (Azadeh et al. 2015). In specific, dis- tribution process, along with transmission operations, must be the center of addressing overall performance efficiency (Tavassoli et al. 2015). Assessing its performance help reg- ulators in planning for improvements to achieve customer satisfaction as well as reduced cost. Furthermore, assessing the significant, specific area of the network is an effective way to clearly look into the sources of technical or operational inefficiencies that would affect the quality of service. In spe- cific, the direct advantage in the efficiency measurement is to deliver quality service in terms of supply, minimum power losses and cost; its indirect advantage is that the continuity of supply of electricity will be improved, resulting in minimized customers’ complaints. Several efficiency measurements are employed in the domain literature that relates to how EDUs can overcome inefficiency issues. These methodologies include (1) stochas- tic frontier panel data model to analyze cost efficiency (Filippini and Wetzel 2014), (2) model predictive conditions (MPC) for geographical load balancing in terms of elec- tricity cost, renewable energy integration, and the number of server switching (Paul et al. 2016), (3) stochastic fron- 123