Machine Learning for Predicting Environmental Impact in Green Buildings: A Systematic Review
Aleke Christiana Ukamaka *
The Federal Polytechnic, Bida, Niger State, Nigeria.
Adepeju Nafisat Sanusi
Catholic University of America, Washington DC, USA.
Joshua Babatunde Asere
Indiana University Bloomington, Indiana, USA.
Hussein Kehinde Sanusi
Sheffield Hallam University, United Kingdom.
*Author to whom correspondence should be addressed.
Abstract
The construction industry significantly contributes to global environmental challenges, accounting for approximately 40% of global energy consumption and 36% of CO2 emissions. Green building practices have emerged as a critical solution, yet accurately predicting their environmental impact remains challenging. This systematic review examines the application of machine learning (ML) techniques for predicting environmental impacts in green buildings. A comprehensive literature search identified 32 relevant studies published between 2018-2024, focusing on energy consumption prediction, carbon footprint assessment, indoor environmental quality, and lifecycle impact analysis. The findings reveal that ensemble methods, deep learning algorithms, and hybrid models demonstrate superior performance in predicting various environmental metrics. Random Forest, Support Vector Machines, and Artificial Neural Networks emerged as the most frequently employed techniques, achieving accuracy rates exceeding 80% in energy consumption predictions. Key challenges include data quality, model interpretability, and integration with building information modeling systems. This review provides insights for researchers, practitioners, and policymakers seeking to leverage ML for sustainable building design and operation.
Keywords: Machine learning, green buildings, environmental impact, energy prediction, sustainability, building performance