دانلود رایگان مقاله اندازه گیری بهره وری انتشار کربن و انرژی منطقه ای

Abstract

The use of data envelopment analysis (DEA) in China’s regional energy efficiency and carbon emission efficiency (EE&CE) assessment has received increasing attention in recent years. This paper conducted a comprehensive survey of empirical studies published in 2006–2015 on China’s regional EE&CE assessment using DEA-type models. The main features used in previous studies were identified, and then the methodological framework for deriving the EE&CE indicators as well as six widely used DEA models were introduced. These DEA models were compared and applied to measure China’s regional EE&CE in 30 provinces/regions between 1995 and 2012. The empirical study indicates that China’s regional EE&CE remained stable in the 9th Five Year Plan (1996–2000), then decreased in the 10th Five Year Plan (2000–2005), and increased a bit in the 11th Five Year Plan (2006–2010). The east region of China had the highest EE&CE while the central area had the lowest. By way of conclusion, some useful points relating to model selection are summarized from both methodological and empirical aspects.

5. Conclusions

This paper provides a comprehensive survey of the studies published on China’s regional energy and carbon emission efficiency (EE&CE) assessment using DEA-type models. It contributes to the literature in the following ways: (1) it is the first comprehensive review of DEA studies on China’s regional EE&CE assessment; (2) it provides a helpful reference for model selection (only DEA models) from both theoretical and practical perspectives; (3) it investigates China’s regional EE&CE using six types of DEA models based on the same data sets for China’s 9th, 10th and 11th Five Year Plans, providing comparable empirical results from different models. A total of six types of DEA models were employed in the previous studies, namely the Radial, M-Radial, RMM, SBMT, RAM and DDF models. The major difference among these models is the way they adjust the inefficiency units. Each model has its application scopes, advantages and limits. From theoretical perspective, Radial model is the simplest and most popular model, because it proportionally adjust the inputs/outputs to the efficient targets. The other models can adjust inputs/outputs non-proportionally. On the other hand, input-/output-oriented should be presupposed for the Radial, M-Radial and RMM models, but is no necessary for the SBMT, RAM and DDF models whose inputs/outputs can be improved simultaneously using slacks or directional vectors. Moreover, the SBMT, RAM and DDF models can provide information about the amount of increase/decrease in inputs/outputs to the efficient targets. From practical perspective, empirical results show evidences that SBMT and DDF models have stronger discrimination power than others in China’s regional EE&CE assessment. According to the aim of different studies, above conclusions provide useful reference for model selection.