Integrating Design Management Practices to Improve Buildability in Construction Projects
Main Article Content
Abstract
The integration of design management practices plays an important role in improving the buildability of building projects. The main focus of this study is on the design decisions, including the task time, resource allocation, and the material choices, in terms of the time and the cost during project development. Data on the quantitative design are used, with machine learning (Random Forest (RF), Support Vector Machines (SVM) and Gradient Boosting (GBM) as well as Monte Carlo Simulations (MCS) to identify the risks and predict the buildability outcome. The research suggests the most significant factors affecting project delay are the task duration and the cost. The Random Forest model has the highest predictive ability (R² = 0.67) with regard to predicting the buildability. The findings emphasise the importance of earlystage design management to minimise risks, improve time schedule fulfilment and reduce costs. The application of modern data analysis techniques will help practitioners make reasonable design decisions that will lead to better buildability and project success. This research also contributes to the body of knowledge in construction project management as it will offer evidence that shows the value of data-driven solutions to improve buildability.