The Evolution of Industrial Symbiosis Reexamined in the Network of Industry Space

Industrial symbiosis (IS) has been regarded as an effective method to reduce environmental pollution and resource consumption, which is especially important in developing countries. However, few IS researches have noticed the... [ view full abstract ]

Industrial symbiosis (IS) has been regarded as an effective method to reduce environmental pollution and resource consumption, which is especially important in developing countries. However, few IS researches have noticed the rapid industrialization process in developing countries, or the deindustrialization in developed countries. The potential of IS should be reexamined based on the fact that industrial structure is evolving and companies enter or exit a region from time to time. In this study, we first considered the co-existence probability of different industries in the same region by calculating a proximity metric, transformed the relationship into a complex network named “industry space”, and identified the complexity of each industry and each region with the network. We used China Industrial Enterprises Database (1998-2009) as our data source, which includes 200000-400000 largest enterprises from 525 industries in China each year. It was found that industrial structure became more complex in most regions during 1998-2009 in China, and the evolution was generally path-dependent: a region would probably develop a new industry proximate to its current industries in the network of industry space. We further examined some famous cases of eco-industrial parks published in journals and summarized about 120 typical IS relations between different industries, which concentrated in several communities (especially the energy-intensive ones) in the network of industry space. Different developing paths lead to great differences in the number, stability and effects of IS relations, which are revealed by both historical data and our simulations. It was shown that China’s 108 demonstration eco-industrial parks had a significantly higher number of typical IS than other regions, and the number of IS had a significantly higher increasing speed during 1998-2009, owing to their increasing regional complexity. However, the regions in the fringe or with a fragmented landscape of industry space could hardly maintain a high IS potential. Our simulations also indicated the IS network of a region might collapse in the future if some keystone yet less complex industries (such as cement) were expelled during the industrial structure transformation. The results suggest that policy makers should take the evolution of industrial structure into consideration when planning or facilitating IS development, and the industry space method may become a useful decision-support tool for its tracking and predicting potential.