AN EXPLAINABLE ARTIFICIAL INTELLIGENCE–ENABLED BIMCENTRIC DIGITAL TWIN FRAMEWORK FOR REAL-TIME CONSTRUCTION MONITORING

The construction industry has been rapidly digitalized, and thus, the process of the use of Building Information Modelling (BIM) and Digital Twin technology has increased in the number of projects being monitored and controlled. Nevertheless, traditional instances of digital twins can be described as visualization platforms that have a short analytical intelligence and lack decision transparency. This paper presents an Explainable Artificial Intelligence-powered BIM- based Digital Twin system of real-time construction monitoring. The scheme combines input data of multi-source construction, such as site progress, IoT sensor feeds, and scheduling in a synchronized digital twin environment. There is a predictive analytics engine built on AI, which predicts delays, safety issues, and productivity variability, as well as an explainable decision interpretation that increases the transparency of the autopilot analysis and recommendations. It is also backed by interactive visualization and decision support interface to help the project stakeholders in proactive planning and mitigation of risks. The results of implementation indicate that there are better monitoring accuracies, quick site dynamics, and increased interpretability in project control procedures. The paper confirms that incorporateability of explainable intelligence into digital twins, generated through BIM, can enhance data-based decision-making, responsibility and operational performance of complex civil infrastructure developments to a large extent.