<![CDATA[Accurate material planning plays a critical role in improving the efficiency of precast bridge construction projects. Inaccurate estimation of material requirements may result in resource waste, production delays, erection disruptions, and increased project costs. Conventional material planning approaches mainly rely on deterministic calculations and practitioners’ experience, making them less effective in handling complex project characteristics and uncertainties. Recently, Random Forest has emerged as one of the most widely applied machine learning algorithms in construction management because of its capability to model nonlinear relationships, process high-dimensional datasets, reduce overfitting, and identify influential variables affecting prediction performance. This study aims to systematically review previous research on the application of Random Forest in construction material planning, with particular emphasis on precast bridge projects. A Systematic Literature Review (SLR) based on the PRISMA 2020 framework was conducted using publications indexed in Scopus, Web of Science, ScienceDirect, SpringerLink, Taylor & Francis, and IEEE Xplore between 2015 and 2025. The review reveals that Random Forest has been successfully implemented for construction cost estimation, resource forecasting, supply chain optimization, productivity prediction, and project risk management. However, its application to material planning in precast bridge construction remains limited. Based on the identified research gap, this study proposes a conceptual framework integrating Random Forest with a Decision Support System (DSS) to support more accurate, adaptive, and data-driven material planning. The findings contribute to advancing artificial intelligence applications in construction management while promoting efficiency and sustainability in precast bridge projects.]]>
Copyrights © 2026
