The construction industry continues to experience significant schedule overruns, with global studies reporting that more than 70% of large-scale construction projects exceed planned timelines. Traditional scheduling techniques, such as the Critical Path Method (CPM) and Program Evaluation and Review Technique (PERT), often fail to capture the dynamic and uncertain nature of construction environments. Recent advances in predictive analytics and machine learning (ML) offer promising opportunities for improving schedule forecasting, risk identification, and decision-making. This paper presents a comprehensive review and conceptual framework for predictive analytics in construction scheduling using ML algorithms. A systematic literature review based on the PRISMA methodology was conducted, examining 30 significant studies published between 2015 and 2026 from Scopus- and Web of Science-indexed sources. The review categorizes existing research into four thematic areas: schedule delay prediction, resource allocation optimization, real-time monitoring and forecasting, and hybrid artificial intelligence approaches. Comparative analysis reveals that ensemble learning methods, particularly Random Forest and XGBoost, consistently outperform traditional statistical models in predicting schedule deviations, achieving accuracy levels exceeding 85% in several studies. Deep learning models demonstrate superior performance when large-scale sensor and Building Information Modeling (BIM) data are available. However, challenges remain regarding data quality, model interpretability, integration with existing project management systems, and scalability across diverse project contexts. The paper proposes a conceptual framework integrating BIM, Internet of Things (IoT), and ML-based predictive analytics for proactive schedule management. The study contributes theoretically by synthesizing fragmented literature, managerially by identifying implementation strategies, technologically by outlining an integrated predictive analytics architecture, and sustainably by demonstrating how improved scheduling can reduce resource waste and carbon emissions. Future research directions include explainable AI, digital twin integration, federated learning, and sustainability-aware scheduling models.

Keywords

Keywords: predictive analytics; construction scheduling; machine learning; schedule delay prediction; BIM; project management; artificial intelligence