Construction costs are a fundamental aspect of the construction industry, encompassing expenses associated with materials, labor, equipment and overheads required to complete a project. Effective cost management is vital for ensuring project feasibility, profitability and the efficient allocation of resources. This research aims to address these challenges by developing a machine learning (ML)-based construction cost prediction model and optimization design. The proposed framework utilizes the Artificial Gorilla Troops optimizer-tuned Intelligent Random Forest (AGT-INT-RF) model to enhance prediction accuracy and reliability. The dataset consists of diverse construction projects, including information on material quantities, labor costs and project timelines. To ensure high data quality, a robust pre-processing pipeline was implemented using outlier detection and normalization techniques. Key features influencing construction costs were extracted using Principal Component Analysis (PCA) to reduce dimensionality and improve computational efficiency. The AGT-INT-RF algorithm optimizes the hyperparameters of the INT-RF through a nature-inspired optimization strategy, improving predictive accuracy while maintaining computational efficiency. This approach predicts construction costs based on critical factors and supports optimization by identifying cost-saving opportunities. In a comparative analysis, the suggested method is assessed with various evaluation measures, such as R squared (0.97), MAPE (0.36), RMSE (0.068) and accuracy (96.85%). Experimental results demonstrate that the AGT-INT-RF model achieves superior performance compared to traditional methods, with high prediction accuracy and reduced error rates. This framework supports construction enterprises in managing costs effectively, optimizing resource allocation, and enhancing risk assessment. By advancing the digitalization and intelligent management of construction projects, and it contributes significantly to the sustainable development of the construction industry.
Zhang et al. (Wed,) studied this question.