A classification method for making-do waste using Machine Learning

Abstract This research study investigates the potential use and best-fitting model for the automated making-do waste classification in construction sites, using Machine Learning techniques to reduce labor and inconsistencies of the manual method. Given the difficulty of manually analyzing a textual database of non-conformities, an automated method applying Machine Learning algorithms is proposed. A total of 8,196 records were collected from the Melius Qualidade service management platform, covering twenty-one high-end multifamily residential projects from three construction companies in Goiânia/GO, of which 3,598 were considered suitable for this research after filtering. The initial classification was done manually, followed by applying nine Machine Learning algorithms by using the Orange Data Mining software for testing and evaluation. Results indicated that grouping data by company yielded the best prediction accuracy, while the Neural Network model achieved a recall of up to 98.20%, making it the most effective. The study highlights that automation accelerates the classification process and improves precision and consistency in identifying making-do waste, significantly contrib