Machine learning for the prediction of rework in green buildings

Green buildings were developed to reduce the negative impact of construction on occupants, communities, and the environment, while also improving contractor productivity and profit margins. However, construction companies have documented instances of rework in green buildigs, despite the advantages and potential of sustainable building designs. This paper reports research that identified the prediction of rework green buildings; Eight machine learning models were trained to predict the occurrence of reworks in green buildings. The dataset consisted of 65 responses and 16 features. Feature scaling and normalisation to standardise the features were performed across the dataset. The research finds variation in the project, cost overrun, time over run, client experience, types of client, specification and project's size were the main predictors of reworks in the green buildings. The results revealed that Gradient Boosting, AdaBoost and Random Forest delivered the most accurate results. The Gradient Boosting model achieved an accuracy of 97% on the test dataset, with a precision of 97%, recall of 97%, and F1-score of 97%. The AdaBoost model achieved a slightly higher accuracy of 97%, with a precision of 97%, recall of 92%, and F1-score of 97%. The research will serve as a valuable reference for all stakeholders involved in the procurement of green buildings. The findings will have significant theoretical and practical implications, guiding future research on the creation of a sustainable built environment