Energy efficiency in Korea: analysis using a hybrid DEA model

Abstract Given the increasing importance of energy efficiency, we appraise the situation in Korea by analyzing past energy consumption patterns. By using a modified hybrid model of index decomposition analysis, artificial neural network (ANN), and data envelopment analysis (DEA), we predict the optimal energy consumption and estimate the difference between the optimal and real values. We decompose primary energy consumption considering the energy loss in a transformation, correcting the over-fitting problem in ANN, and addressing the negative value issue in DEA. We find that energy consumption was the most efficient between 1993 and 1994, 1994 and 1995, 1997 and 1998, and 1999 and 2000. If the over-fitting and negative value problems are properly controlled, the presented LMDI-ANN-DEA hybrid model can be used to predict energy efficiency. The results of this study would be useful to analyze energy consumption patterns in the benchmark years of Korea and help the government formulate a suitable energy policy.

[1]  B. W. Ang,et al.  Decomposition of aggregate energy intensity changes in two measures: ratio and difference , 2003 .

[2]  B. W. Ang,et al.  The LMDI approach to decomposition analysis: a practical guide , 2005 .

[3]  John C. B. Cooper.  Artificial neural networks versus multivariate statistics: An application from economics , 1999 .

[4]  B. W. Ang,et al.  Accounting frameworks for tracking energy efficiency trends , 2010 .

[5]  Salvatore Manco,et al.  Energy efficiency measurement in industrial processes , 2012 .

[6]  B. W. Ang,et al.  Factorizing changes in energy and environmental indicators through decomposition , 1998 .

[7]  P. Linares,et al.  Energy Efficiency: Economics and Policy , 2010 .

[8]  E. Mine Cinar,et al.  Neural Networks: A New Tool for Predicting Thrift Failures , 1992 .

[9]  A. U.S.,et al.  Measuring the efficiency of decision making units , 2003 .

[10]  Jinsoo Kim,et al.  Forecasting Natural Gas Prices Using Wavelets, Time Series, and Artificial Neural Networks , 2015, PloS one.

[11]  Aie,et al.  World Energy Outlook 2013 , 2013 .

[12]  O. A. Olanrewaju,et al.  Assessing the energy potential in the South African industry: A combined IDA-ANN-DEA (Index Decomposition Analysis-Artificial Neural Network-Data Envelopment Analysis) model , 2013 .

[13]  Juan Aparicio,et al.  Translation Invariance in Data Envelopment Analysis , 2015 .

[14]  Energy and economic growth: causality analysis using decomposed energy consumption , 2012 .

[15]  B. W. Ang,et al.  Decomposition of the energy-intensity index with application for the Korean manufacturing industry , 1995 .

[16]  Emmanuel Thanassoulis,et al.  Negative data in DEA: a directional distance approach applied to bank branches , 2004, J. Oper. Res. Soc..

[17]  Jesús T. Pastor,et al.  Chapter 3 Translation invariance in data envelopment analysis: A generalization , 1996, Ann. Oper. Res..

[18]  A. Charnes,et al.  Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis , 1984 .

[19]  Yehoshua Y. Zeevi,et al.  Neural networks: theory and applications , 1992 .

[20]  Pierre Desprairies,et al.  World Energy Outlook , 1977 .

[21]  O. A. Olanrewaju,et al.  Integrated IDA–ANN–DEA for assessment and optimization of energy consumption in industrial sectors , 2012 .

[22]  Joe Zhu,et al.  Evaluating Hedge Fund and CTA Performance: Data Envelopment Analysis Approach , 2005 .

[23]  S. Borenstein.  A Microeconomic Framework for Evaluating Energy Efficiency Rebound and Some Implications , 2013 .