Predictive Scheduling for Repetitive Construction: A Machine Learning Framework for Enhanced Resource and Time Management

Repetitive construction projects, characterized by the sequential execution of similar tasks across multiple units, present unique scheduling challenges, particularly concerning resource crew continuity and overall project duration. Traditional scheduling methods often fall short in optimizing these dual objectives simultaneously. This paper proposes a novel framework that integrates machine learning (ML) algorithms with established repetitive project scheduling principles to achieve enhanced optimal scheduling. Various ML models are applied to predict optimal crew allocation strategies and forecast potential work breaks or delays, thereby informing dynamic scheduling adjustments. The methodology aims to balance the minimization of project duration with the maximization of crew work continuity, a critical factor for productivity and cost efficiency. Through a comprehensive evaluation of ML model performance metrics, we demonstrate the efficacy of this data-driven approach in providing predictive insights and decision support for project managers. The findings highlight the transformative potential of machine learning in revolutionizing repetitive construction project scheduling, leading to more efficient resource utilization, reduced project risks, and improved overall project performance.