**Using BIM Technology and Machine Learning to Predict Delays in Construction Projects**
**Using BIM Technology and Machine Learning to Predict Delays in Construction Projects**
BAHAR SALARI1
1) B.Sc. in Civil Engineering
IOI Code
:
ABKD-ACZFKD
Publication :
Fourth International Conference on applied researches in civil engineering, architecture and urban planning (icarcau.com)
Abstract :
Delays in the execution of construction projects are among the most costly and common challenges facing the construction industry, with direct impacts on cost, quality, and stakeholder satisfaction. In recent years, the integration of Building Information Modeling (BIM) with machine learning methods has emerged as a novel approach for the early prediction of delays and proactive project management. In this paper, through an analytical review of five studies published in the areas of BIM-based sustainability assessment, educational perceptions of cost overrun management, resource efficiency optimization using a hybrid ANN-SEM approach, progress management based on reinforcement learning and digital twin technologies, and barriers to the implementation of intelligent chatbots in BIM libraries, a conceptual framework for predicting delays in construction projects is proposed. The proposed framework employs four-dimensional BIM scheduling data, cost and resource records, and environmental variables as inputs to machine learning algorithms, including Artificial Neural Networks (ANN), Random Forest (RF), Support Vector Machines (SVM), and XGBoost, as well as reinforcement learning, to identify potential project time deviations at an early stage. The findings from the literature review indicate that hybrid models, particularly gradient-boosting-based approaches and reinforcement learning integrated with digital twins, provide higher prediction accuracy and greater adaptability to dynamic disruptions in construction processes. Finally, implementation challenges, including data quality and integration, users’ technical knowledge and skills, and the need for specialized training, are discussed, and recommendations for future research are provided.
**Keywords:** Building Information Modeling (BIM); Machine Learning; Project Delay Prediction; Reinforcement Learning; Digital Twin; Construction Project Management