Data Science Applications in Circular Economy: Trends, Status, and Future.

The circular economy (CE) aims to decouple the growth of the economy from the consumption of finite resources through strategies, such as eliminating waste, circulating materials in use, and regenerating natural systems. Due to the rapid development of data science (DS), promising progress has been made in the transition toward CE in the past decade. DS offers various methods to achieve accurate predictions, accelerate product sustainable design, prolong asset life, optimize the infrastructure needed to circulate materials, and provide evidence-based insights. Despite the exciting scientific advances in this field, there still lacks a comprehensive review on this topic to summarize past achievements, synthesize knowledge gained, and navigate future research directions. In this paper, we try to summarize how DS accelerated the transition to CE. We conducted a critical review of where and how DS has helped the CE transition with a focus on four areas including (1) characterizing socioeconomic metabolism, (2) reducing unnecessary waste generation by enhancing material efficiency and optimizing product design, (3) extending product lifetime through repair, and (4) facilitating waste reuse and recycling. We also introduced the limitations and challenges in the current applications and discussed opportunities to provide a clear roadmap for future research in this field.

Analyzing the socioeconomic determinants of PM2.5 air pollution at the global level.

Since PM2.5 pollution has jeopardized public health, the research on how ambient fine particulate matter (PM2.5) concentrations are influenced has been increasingly important for the implementation of regional PM2.5 concentration reduction. This study analyzed the socioeconomic determinants of PM2.5 air pollution of 132 countries/economies. It was found that the main inhibitor for the PM2.5 air pollution is the emission intensity (EmI), which is measured by the PM2.5 emission when a united of energy is consumed, in every income level of countries, while the energy intensity (EnI) is the second inhibitor. Meanwhile, economic output (EO) was the largest driving factor on the PM2.5 concentrations, while population (P) growth was the second. Overall, the national employment rate (Emp) showed very little impact on the countries. This study also analyzed the income-based variation in the effects of the five factors on PM2.5 concentration changes: overall, the effects of the determinants all decreased with the rise of income levels, i.e., both the inhibiting effects of PM2.5 EmI and EnI and driving effects of EO and P performed stronger in lower-income countries than higher-income ones. Regarding the income-based variation of the determinants, this study also discussed the policy implications, such as adopting technologies on reducing PM2.5 intensity and EnI, transferring the EO from the manufacturing industry to the service industry, and international organizations on public health and environmental protection should provide targeted strategies, guidelines, and other assistances to lower-income countries as both driving and inhibiting factors performed more influential on their PM2.5 concentration changes.

Using Deep Learning to Fill Data Gaps in Environmental Footprint Accounting.

Environmental footprint accounting relies on economic input-output (IO) models. However, the compilation of IO models is costly and time-consuming, leading to the lack of timely detailed IO data. The RAS method is traditionally used to predict future IO tables but suffers from doubts for unreliable estimations. Here we develop a machine learning-augmented method to improve the accuracy of the prediction of IO tables using the US summary-level tables as a demonstration. The model is constructed by combining the RAS method with a deep neural network (DNN) model in which the RAS method provides a baseline prediction and the DNN model makes further improvements on the areas where RAS tended to have poor performance. Our results show that the DNN model can significantly improve the performance on those areas in IO tables for short-term prediction (one year) where RAS alone has poor performance, R2 improved from 0.6412 to 0.8726, and median APE decreased from 37.49% to 11.35%. For long-term prediction (5 years), the improvements are even more significant where the R2 is improved from 0.5271 to 0.7893 and median average percentage error is decreased from 51.12% to 18.26%. Our case study on evaluating the US carbon footprint accounts based on the estimated IO table also demonstrates the applicability of the model. Our method can help generate timely IO tables to provide fundamental data for a variety of environmental footprint analyses.

Estimation of Unit Process Data for Life Cycle Assessment Using a Decision Tree-Based Approach.

Lacking unit process data is a major challenge for developing life cycle inventory (LCI) in life cycle assessment (LCA). Previously, we developed a similarity-based approach to estimate missing unit process data, which works only when less than 5% of the data are missing in a unit process. In this study, we developed a more flexible machine learning model to estimate missing unit process data as a complement to our previous method. In particular, we adopted a decision tree-based supervised learning approach to use an existing unit process dataset (ecoinvent 3.1) to characterize the relationship between the known information (predictors) and the missing one (response). The results show that our model can successfully classify the zero and nonzero flows with a very low misclassification rate (0.79% when 10% of the data are missing). For nonzero flows, the model can accurately estimate their values with an R2 over 0.7 when less than 20% of data are missing in one unit process. Our method can provide important data to complement primary LCI data for LCA studies and demonstrates the promising applications of machine learning techniques in LCA.

Urban Air Pollution Mapping Using Fleet Vehicles as Mobile Monitors and Machine Learning.

Spatially explicit urban air quality information is important for developing effective air quality control measures. Traditionally, urban air quality is measured by networks of stationary monitors that are not universally available and sparsely sited. Mobile air quality monitoring using equipped vehicles is a promising alternative but has focused on vehicle-level experiments and lacks fleet-level demonstration. Here, we equipped 260 electric vehicles in a ride-hailing fleet in Beijing, China with low-cost sensors to collect real-time, spatial-resolved data on fine particulate matter (PM2.5) concentrations. Using this data, we developed a decision tree model to infer the distribution of PM2.5 concentrations in Beijing at 1 km by 1 km and 1 h resolution. Our results are able to show both short- and long-term variations of urban PM2.5 concentrations and identify local air pollution hotspots. Compared with a benchmark model that only uses data from stationary monitoring sits, our model has shown significant improvement with the coefficient of determination increased from 0.56 to 0.80 and root mean square error decreased from 12.6 to 8.1 µg/m3. To the best of our knowledge, this study collects the largest mobile sensor data for urban air quality monitoring, which are augmented by state-of-the-art machine learning techniques to derive high-quality urban air pollution mapping. Our results demonstrate the potential and necessity of using fleet vehicles as routine mobile sensors combined with advanced data science methods to provide high-resolution urban air quality monitoring.

Analysis on the carbon emission peaks of China's industrial, building, transport, and agricultural sectors.

Carbon emission peak has become a focus of political and academic concern in global community since the launch of Kyoto Protocol. China, as the largest carbon emitter, has committed to reaching the carbon peak by 2030 in Paris Agreement. This ambitious national goal requires the endeavors of individual sectors, particularly those carbon-intensive ones. Predicting the sectoral peaks under current endeavors and understanding driving forces for the carbon emission changes in the past years are substantial for guiding the allocation of the country's future efforts. In the past studies contextualized in China, the prediction of its carbon peaks seldom appeared at the sectoral level, which is considered as a research gap. Therefore, this study predicts the peaks at four carbon pillar sectors (i.e. industrial, building, transport and agricultural sectors) and identifies the driving forces for the carbon emission changes of them. This study hypothesized Carbon Kuznets curve (CKC) as the theoretical model for predicting the peaks and used Logarithmic mean Divisia index (LMDI) as the method to identify the driving forces. The results show that the carbon emission in the country will peak in 2036, six years later than the agreed year. The lateness of the national peak can be attributed to the significant lateness of three pillar sectors' peaks, occurring in 2031 for the industrial sector, 2035 for the building sector, 2043 for the transport sector, peak for the agricultural sector occurs four years earlier in 2026 though. Furthermore, the results show that carbon emission is significantly driven by the booming economic output and inhibited by decreasing energy intensity, but the slight fluctuation of energy structure plays a minor role in the four sectors. Policy adjustments are proposed for effectively and efficiently urging the on-time occurrence of the national peak.

A three-step strategy for decoupling economic growth from carbon emission: Empirical evidences from 133 countries.

To decouple the economic growth and carbon emission has been considered imperative to promote low-carbon economy. Nevertheless, previous studies on decoupling analysis between economic growth and carbon emission were contextualized merely in individual countries rather than the globe, which are insufficient for developing the low-carbon economy as a global target. Carbon intensity (CI), carbon emission per capita (CP), and total carbon emission (TC) serve as three important indicators of the status of regional carbon emission, but only decoupling economic growth from TC was analyzed in previous studies. To close the two gaps, this study aims to investigate the global decoupling statuses of economic growth from not only TC but also CI and CP by using Tapio decoupling index. The decoupling statuses of 133 countries and income-level groups to which they are classified are identified using the data from 2000 to 2014. According to the results, it is observed that economic growth decouples from CI, CP, and TC in sequential order, which is called three-step decoupling. In the period, countries whose economic growth having decoupled from CI, CP, and TC, account for 74%, 35% and 21% respectively. Higher income-level group has the larger proportion of countries having reached their decoupling statuses. These findings may serve as valuable references for policy-makers to understand the current decoupling statuses and make three-step strategies if necessary towards the global low-carbon economy.

What are the root causes hindering the implementation of green roofs in urban China?

It is worldwide accepted that green roofs have a variety of environmental, economic, and social benefits. China, which is experiencing rapid urbanization, has great potential to gain the benefits of green roofs, yet which are not commonly seen in the existing or new buildings. Understanding its root causes is important for promoting the larger-scale implementation of green roofs. Previous studies have studied the barriers of implementing green roofs in developed urban areas but ignored developing countries or regions, whose implementation of green roofs is still at the initial stage. To fill the research gap, this study aims to investigate the root causes that impede the implementation of green roofs in urban China through a practical survey and case study. The root causes are identified as the increase of maintenance cost, increase of design and construction cost, poor arrangement of the use of green roofs, and lack of incentives towards developers. Policy implications are proposed, which provide valuable references for decision-makers to improve the green-roof-related codes, policies and incentives.

Decoupling China's economic growth from carbon emissions: Empirical studies from 30 Chinese provinces (2001-2015).

The world has witnessed unparalleled economic development over the past decades, but accompanied by large amount of carbon emissions, which triggered the global warming. It is critical for the global sustainable development by decoupling economic growth from carbon emissions at country level, specifically for the largest emitter, China. This study conducts a decoupling analysis from the perspective of carbon intensity (CI), per capita carbon emissions (PC) and total carbon emissions (TC) with reference to 30 Chinese provinces, covering the period of 2001-2015. Based on the Log Mean Divisa Index (LMDI) method, the effects of energy structure (ES), energy intensity (EI), economic output (EO) and population size (P) on TC at provincial level are investigated. Results show that: (1) a strong decoupling relation between GDP and CI is found in most provinces except Hainan, Qinghai and Xinjiang, while there is large room for China to decouple completely from PC and TC; (2) EO and EI are the dominated inhibiting and promoting factors respectively for carbon emission reduction; (3) ES effect on increasing carbon emission changes between positive and negative, while P has a positive but insignificant effect on the increase of carbon emissions for most provinces. The results help local governments formulate measures to coordinate regional economic development and carbon emission reduction.

Critical Success Factors (CSFs) for the Adaptive Reuse of Industrial Buildings in Hong Kong.

With the economic restructuring during the 1980s and 1990s in Hong Kong, most manufacturing plants were relocated to China and many industrial buildings were left neglected or vacant. At the same time, owing to limited land supply, a shortage of affordable housing has been a problem in Hong Kong for many years. Adaptive reuse of industrial buildings may be a way of solving this problem. However, adaptive reuse is not an easy decision because there are many factors affecting adaptive reuse. Therefore, this paper examines the current situation of adaptive reuse of industrial buildings in Hong Kong and identifies a list of factors affecting the adaptive reuse of industrial buildings. Six factors are considered Critical Success Factors (CSFs). Based on a Principal Component Analysis, 33 factors are grouped into eight principal components, namely, sustainability, economics and finance, the market, changeability, location and neighborhood, culture and public interests, legal and regulatory matters, and the physical condition of the building. The identified CSFs and principal factors provide a useful reference for various stakeholders to have a clear understanding of the adaptive reuse of industrial buildings in Hong Kong, especially for the government to review current policies of adaptive reuse.