Accelerating Whole Life Carbon Assessment in Construction with Artificial Intelligence

Abstract

The global effort to reduce carbon emission and mitigate the environmental carbon footprint of the construction industry, along with its impact on climate change, has prompted construction organisations to integrate life cycle carbon assessment into their practices. One of the key areas for enhancing sustainability is the immediate evaluation of carbon footprint in the design stage of construction projects. This includes carbon emissions associated with the intrinsic properties of materials, as well as those related to transportation and installation. Additionally, there are carbon emissions linked to the maintenance and operation of the built asset throughout a project life cycle. This paper aims to accelerate whole-life carbon assessment by integrating artificial intelligence with CarboniCa software, an in-house carbon assessment tool utilised by a major UK construction organisation. To speed up the evaluation process, a new development is suggested using AI deep-learning neural networks to learn from experience and to estimate carbon footprint, thus reducing time, energy and cost. By leveraging historical construction project data within the CarboniCa software, the experimental results provided a reasonable estimation (R² = 0.87) of the whole-life carbon for different building types. With the integration of deep learning neural networks, the proposed process is expected to improve efficiency by saving time and resources. It will provide designers with rapid guidance during the early design stage, enabling them to reduce the life-cycle carbon impact more effectively. The paper begins with a literature review on the significance of life cycle carbon assessment in the construction industry, followed by an overview of CarboniCa, a carbon assessment tool. It then explores the integration of artificial intelligence to enhance the software’s ability to rapidly evaluate whole-life carbon, thereby promoting sustainability within the built environment.