Recently, the RGC has announced the results of the National Natural Science Foundation of China (NSFC) / Research Grants Council (RGC) Joint Research Scheme 2025/26 (https://www.ugc.edu.hk/eng/rgc/funding_opport/nsfc/funded_research/list_award_e2025.html). The collaborative research project, entitled “Multivariate assessment, AI-supported prediction, and systematic reduction of lifecycle carbon emissions of the building stock of high-density cities”, led by Prof. Wei Pan from The University of Hong Kong and Prof. Da Yan from Tsinghua University, has been awarded funding of HK$1,178,000 (Project Number: N_HKU764/25; NSFC Grant Number: 52561160148).

​​

​

Project Summary

Energy saving and carbon reduction are critical and urgent in Hong Kong and the Chinese Mainland to meet “dual carbon” goals of peaking emissions by 2030 and achieving neutrality by 2050/2060 while supporting the global goal of keeping warming below 1.5°C by 2100. The carbon emission of the building industry causes 37% of the global carbon emissions, so it is necessary to analyze and mitigate lifecycle carbon emissions (LCCE). City-level LCCE analysis is important, as the building stock significantly impacts city carbon output. There are two main methods to evaluate lifecycle carbon assessment (LCCA): bottom-up and top-down. Bottom-up models integrate with detailed data on individual buildings and are great for testing specific technologies or policies, but are constrained by high data requirements and complex modeling processes. Top-down methods use broader, easier-to-get data but fail to break down emissions by activity or source, making it hard to plan future fixes on policy or technology. City-level LCCA for the building stock faces challenges: 1) insufficient integration of multiple factors, with complexities and uncertainties arising from scaling LCCA to urban levels; 2) difficulties in identifying critical parameters influencing LCCE in high-density urban contexts, coupled with limited understanding of data dynamics, complexity, uncertainty, and long-term impacts; and 3) immature integration of AI techniques with LCCA models, despite AI’s potential to enhance emission prediction, optimization, and climate change mitigation.

This project will address a main research question: How can the lifecycle carbon emissions of the building stock at the city level be comprehensively assessed, accurately predicted and systematically reduced under the “dual carbon” goals in China and the global carbon neutrality target? The project will be carried out over 48 months in three linked tasks: 1) develop and validate a multivariate LCCA model using the combination of top-down and bottom-up methods for the building stock of high-density cities; 2) assess the impact of critical parameters and variables and their relevant data missingness, dynamics, complexity and uncertainty and on the LCCE of the building stock through univariate and multivariate analysis; 3) integrate advanced AI techniques into the assessment model to predict LCCE toward 2050/2100 and develop optimal scenarios and strategies. The project outcomes will greatly advance the knowledge, method and practice of multivariate assessment, AI-supported prediction, and systematic reduction of LCCE of the building stock of high-density cities.

The project management team will include six investigators from The University of Hong Kong, Tsinghua University and Southeast University to cover expertise in life cycle carbon emission analysis, construction management, and green building. They will collaborate with relevant government bureaus, departments and industry organizations.