A modified super-efficiency network data envelopment analysis: Assessing regional sustainability performance in China

Abstract

The super-efficiency technique aims to improve the discriminating power of data envelopment analysis (DEA) through re-evaluation after excluding the evaluated object from the reference set. However, the super-efficiency radial network DEA with undesirable production variables faces the infeasibility problem as the traditional DEA. To solve this issue, this paper detects one new origin of infeasibility in network DEA not existing in traditional DEA and provides the necessary and sufficient conditions of infeasibility occurrence. Then, a modified super-efficiency radial network DEA is developed without infeasibility. We apply the modified model to assess the regional sustainability performance in China from 2009 to 2017. Considering the internal connectivity and influences among the economic, environmental, and social dimensions of sustainability, we construct a three-stage network system, i.e., economic growth (EG), waste disposal (WD), and health protection (HP) subsystems. The results show that the regional sustainability performance of China was on a downward trend until 2015 but has improved significantly since then. Besides, the status quo of a regional disparity in sustainability performance is confirmed. The results suggest that the eastern region remains the leader in promoting the regional sustainability of China, the central and western regions are trying to catch up with it, and the northeast region still needs more support from the government to keep pace with the coordinated development of China.