PREDICTION OF LABOR PRODUCTIVITY IN CONSTRUCTION PROCESSES USING ARTIFICIAL INTELLIGENCE TECHNIQUES

The accurate prediction of labor productivity is critical for achieving efficiency, cost control, and timely completion in construction projects. However, conventional forecasting methods often fail to capture the nonlinear and dynamic interactions among human, environmental, and managerial factors. This study presents an artificial intelligence (AI)-based framework for predicting labor productivity in construction processes using machine learning and deep learning techniques. Three models—Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machine (SVM)—were developed and trained using an integrated dataset containing variables such as labor hours, equipment availability, worker experience, material supply, temperature, and supervisory efficiency. The models were evaluated using the coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE). The ANN model achieved the highest predictive accuracy (R² = 0.94), followed by RF (R² = 0.91) and SVM (R² = 0.88). Residual and feature importance analyses revealed that labor hours (23.6%), equipment availability (18.4%), and worker experience (14.7%) were the most influential parameters affecting productivity. The ANN exhibited a symmetric and unbiased error distribution, confirming its robustness and generalization capability across diverse project conditions. The proposed AI-based framework offers a data-driven decision-support system that enables project managers to proactively monitor performance, allocate resources efficiently, and minimize schedule deviations.