People around the world are moving to cities in order to become wealthy, especially in developing countries undergoing socio-economic transition. This will dramatically change the way that society consumes resources, in terms of both materials and energy. We will need to better understand how resource use changes as development happens in order to focus efforts to reduce that resource use. China is undergoing this transition as it enters the global stage, and with global implications for resource use. The building sector in China is especially resource-intensive, fundamentally driven by the country's urbanization, and plays a prominent role in the Chinese economy, making it a good candidate for the study of how systems adopt sustainable practices. 

This study approaches this problem in two ways -- first using machine learning techniques to analyze time series data from Chinese statistical yearbooks in order to identify and compare transitions across provinces in China. In comparing the evolution of the building sector across provinces, this analysis suggests that long-term development pathways are more evident than discrete development stages. Urbanization levels are the strongest differentiator of these pathways, though other factors, among them geography, are clearly influential.

Fig 1. A machine learning approach, k-means statistical clustering, is used to identify resource use patterns in the Chinese building sector. This graph visualizes the resulting 'clusters' along axes of the underlying data's principal components.

The second approach is qualitative, driven by interviews conducted primarily in Beijing with various stakeholders in the Chinese building sector. This analysis discusses the particularities of the actors and processes involved in Chinese real estate development, identifies opportunities for and barriers to reduced life-cycle energy use in buildings, and describes three distinctive and ongoing sustainability experiments with the potential for significant resource use reductions.

Fig. 2 This diagram represents the underlying institutional structure and stakeholder relationships in China, which have profound implications on resource use in cities.

This research emphasizes the need for integrated approaches to both research and practice in approaching sustainability transitions, and provides a set of complementary frameworks for their analysis.

Fig. 3 This diagram shows how many types of transition between different states of resource use are relatively easily reversible; one may make a distinction between such transitions and larger system transformation, defined here as a transition (or set of complementary transitions) that leads to the much harder to reverse system 'lock-in'.