Data science applications for carbon footprint management in buildings: a systematic literature review

Abstract

Buildings have a significant impact on climate change. The building industry is the world’s biggest energy consumer and the building's operation accounts for 80–90% of its total energy consumption over its lifetime. Data-driven solutions for the management of carbon footprint in buildings have great potential due to the data science field's rapid growth and the expansion of operational building data availability. Therefore, this study's aim is set as to investigate the potential applications of data science for the management of carbon footprint in buildings. The study adopted a systematic literature review as a research methodology. Accordingly, 31 publications were reviewed using the content analysis technique. The study revealed that facilitating pre-process of the operational data of buildings, fault detection and diagnosis, implementing waste management in buildings, conducting the building energy performance modelling, conducting the parametric analysis at the design phase, evaluating the energy efficiency of building designs, benchmarking evaluation, control optimisation and retrofitting analysis are the major applications of data science to the management of carbon footprint in buildings. Moreover, the study suggested carrying more studies should be done on automating and building operational data pre-processing tasks, gathering sufficient labelled data for all possible faulty operations and applying modern big data management tools and advanced analytics techniques lead to improve the applications of data science in the built environment. The results from this study provide better guidance to building sector stakeholders, information technology sector stakeholders, academic persons, non-governmental organisations (NGOs) and other relevant authorities to address the carbon footprint in buildings using data science applications.