This integrative literature review examines the role of machine learning (ML) and
big data analytics in transforming risk management and resource optimization in
construction projects. Following a PRISMA-P-guided search across Scopus,
Google Scholar, Web of Science, and PubMed, 5,756 initial records were
identified and screened, resulting in a final dataset of 154 eligible studies, of which
59 high- and moderate-quality studies were retained for detailed synthesis after
quality assessment. The review highlights ML’s enhanced predictive capabilities
for risk assessment and resource allocation, supported by cross-industry
comparisons and construction-specific case studies. Representative findings from
the literature include reported prediction accuracies of up to 93.75% for delay-risk
prediction and 97.6% for equipment matching, indicating the potential of ML
based tools to improve forecasting, monitoring, and resource-allocation decisions.
The review also highlights the synergistic integration of ML with Building
Information Modeling (BIM), the Internet of Things (IoT), and digital twin
technologies, which collectively enhance project efficiency despite challenges in
data-sharing standardization, interoperability, and regulatory compliance. Key
barriers to ML and big data adoption are identified, along with strategic measures
to address them. The study proposes two novel frameworks for the construction
sector: an AI-Enhanced Construction Risk Prediction and Mitigation Framework
and a Smart Build: AI-Optimized Resource Management Framework. These
frameworks, informed by insights from industry practitioners, policymakers, and
researchers, aim to advance digital transformation in construction by providing
structured approaches for leveraging ML in risk management and resource
optimization.
