Predicting implications of design changes in BIM-based construction projects through machine learning

Abstract

Design changes are not uncommon occurrence in construction projects, which results in implications in several ways including impacts on time and cost. On the other hand, despite Building Information Modelling (BIM) has been around over two decades, its potential has not been much unlocked particularly in connection to predicting implications of design changes. Through a rigorous Design Science Research approach, a conceptual model is outlined in which BIM and ML are integrated. The outlined model is applied to a hypothetical use-case, leading to a tested and validated final conceptual model. The model application demonstrates that the implications of design changes can be predicted much earlier and therefore, time and cost impacts can be better comprehended to form the decision-making. This model can be expanded in future to provide a communal platform where different stakeholders in construction projects can effectively communicate, collaborate, and coordinate particularly in the design changes scenarios.