Enhancing construction management with machine learning-integrated building information modeling: a transformative approach

Machine learning (ML) can significantly enhance Building Information Modeling (BIM) in construction projects by optimizing various processes and decision-making. ML with BIM integration can make construction projects more efficient, safer, cost-effective, and sustainable, while also improving the decision-making process at every stage. However, limited research focused on this integration although ML methods have become popular for forecasting project delays in BIM-enabled building projects recently. This study employs a structural equation model (SEM) to quantify the relationship between possible elements of use of ML and BIM models. It is examined how machine learning algorithms can be used project delay predictions and identified key components to fill the gap. The mixed-methods research used in this study includes a complete literature review, consultation with 10 construction engineering experts for expert perspectives, a preliminary survey, and a main questionnaire survey. Exploratory factor analysis (EFA) identified the key constructs, and structural equation modeling (SEM) validated the connections between the areas of use and project delay forecasts. Technical forecasts, resource optimization, operational suggestions, and project controllability were determined as leading areas from the research. The above structures included risk identification, resource allocation, data-driven decision-making, and progress tracking are possible areas of use of ML. Results indicated that ML has huge potential for delay forecasts, and identified potential areas can serve as potential use areas of ML and BIM integration.