Assessing the effects of the influencing factors on industrial green competitiveness fusing fuzzy C-means, rough set and fuzzy artificial neural network methods

Improving the industrial green competitiveness is of great significance to handle the issues of resource shortage and environmental degradation in the process of ecological civilization construction. The existing literatures on the influencing factors of the industrial green competitiveness mostly examine the linear relationship using econometric models but rarely investigate the nonlinear effect between them. To address this research gap, this study firstly adopts the direction distance function and the global Malmquist-Luenberger productivity index to evaluate the green competitiveness of 33 industrial sub-sectors in China from 2002 to 2017, and employs the fuzzy C-means algorithm to classify these sectors into three categories with high-, medium- and low-levels of green competitiveness, respectively. Furthermore, this study identifies the critical factors influencing the industrial green competitiveness using the rough set theory, and more specifically explores how these critical factors differently impact the green competitiveness of three categories of industrial sub-sectors using the fuzzy artificial neural network. The results indicate that, the green competitiveness of the overall industrial sector in China shows an upward trend during 2002 to 2017, but there exist significant disparities among industrial sub-sectors. Research and development intensity, environmental regulation and enterprise scale are three critical factors influencing industrial green competitiveness. In the sub-industries with high-, medium- and low-levels of green competitiveness, the most significant factor is enterprise scale, research and development intensity, and environmental regulation, respectively.