Abstract
The logistics and manufacturing industries are basic industries that support social development. First, this study classifies the carbon emissions from the logistics industry and pollution emissions from the manufacturing industry as undesirable outputs and evaluates the ecological efficiency of the logistics and manufacturing industries in the Yangtze River Delta from 2006 to 2016 by using the unexpected slacks-based measure (SBM) model. Second, the study analyzes the spatial differences in industrial correlation efficiency by using the exploratory spatial data analysis method. Third, the spatial econometric model is used to analyze the driving factors of the linkage ecological efficiency between logistics industry and manufacturing industry. Finally, the neural network model is used to predict the linkage ecological efficiency. The results show that the ecological efficiency of the manufacturing industry has steadily improved. The ecological efficiency of the logistics industry presents the rising trend in fluctuation. The level of the linkage development between the logistics and manufacturing industries is high. The results of the spatial heterogeneity analysis show that the spatial differentiation of high–high agglomeration and low–low agglomeration is obvious. The spatial agglomeration characteristics are relatively stable, and the spatial diffusion effect is strong. In space, the linkage ecological efficiency shows a trend of development from multiple agglomeration areas to one agglomeration area. The results of driving factor analysis show that foreign direct investment (FDI), government intervention (GI), and human capital (HP) have positive effects on linkage ecological efficiency, while industrial structure (IS), environmental regulation (ER), and energy intensity (EI) have negative effects on linkage ecological efficiency. The results of the linkage development trend analysis show that the linkage ecological efficiency of the two industries will tend to be stable in the future.
Highlights
At present, with the rapid development of the social economy, China’s logistics demand is increasing rapidly, and the scale of logistics is expanding rapidly
The ecological efficiency of logistics and manufacturing industry is evaluated by the slacks-based measure (SBM) model; the spatial heterogeneity and spatial aggregation of the efficiency of the two industries are analyzed by the spatial correlation analysis; the driving factors of linkage efficiency are analyzed by using the spatial measurement method; and the dynamic evolution of the trend is predicted by using a neural network
The calculation results of industrial linkage ecological efficiency show that the ecological efficiency of the manufacturing industry in the study area has steadily improved; the ecological efficiency of the logistics industry has fluctuated greatly, first rising, falling, and steadily improving; and the deviation of the ecological efficiency curve between the logistics and manufacturing industries shows that when considering the undesirable outputs, the logistics and manufacturing industries cannot achieve an absolutely high level of linkage development during the research period, but the ecological efficiency of the logistics and manufacturing industries is relatively stable
Summary
With the rapid development of the social economy, China’s logistics demand is increasing rapidly, and the scale of logistics is expanding rapidly. Slacks-Based Measure Model of Industrial Linkage Ecological Efficiency In actual social production, when people’s material demand is solved, it produces various kinds of side effect products, such as wastewater, waste residue, waste gas, and other pollutants. This article defines these indicators as undesirable outputs. This study selects the super-efficiency SBM model considering unexpected output, which can more comprehensively reflect the concept of environmental efficiency of regional logistics efficiency measurement, including the technical structure relationship between an unexpected output and the factor input. I1 where x is the input variable, xi is the learning sample corresponding to the ith neuron, yij is the connection weight between the pattern layer and the summation layer, and yj is the output result corresponding to the jth neuron in the output layer
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
