Sarah Qutub, Sara Baajajah, Danah Alqalayta, Sadia Samar Ali, Rimal Abutaha, Hani M. Aburas
The rapid expansion of Saudi Arabia's construction sector has intensified environmental press ures, highlighting the urgent need for effective green supply chain management (GSCM) practices. Despite the policy mom entum created by Saudi Vision 2030, limited empirical research explains how GSCM practices influence subcontractors' su stainability performance, particularly from an integrated theoretical and predictive perspective. This study addresses this ga p by integrating institutional theory, the resource-based view (RBV), and the natural resource-based view (NRBV) to examin e the mechanisms through which GSCM practices enhance sustainability performance. Methodologically, the study develop s a hybrid structural equation modelling (SEM)–machine learning (ML) framework, where SEM validates theory-driven caus al relationships and ML improves predictive capability and enables subcontractor sustainability ranking. Using survey data c ollected from construction subcontractors in Saudi Arabia, the SEM results reveal that coercive pressure, information shari ng, sustainable procurement, pollution prevention, and product stewardship significantly influence sustainability performance , while top management support partially mediates these relationships. Complementary ML analysis further identifies the mo st influential GSCM practices driving sustainability outcomes, providing enhanced predictive insights for decision-making. The findings contribute to the literature on sustainable construction supply chains by advancing theoretical integration and met hodological innovation. Practically, the study offers actionable insights for policymakers, contractors, and subcontractors se eking to strengthen sustainability performance in alignment with Saudi Vision 2030 and the Saudi Green Initiative, with broad er implications for emerging economies pursuing sustainable infrastructure development.