A hybrid mathematical model for evaluating recyclability pathways of cement-based materials in Thailand

This study integrates System Dynamics (SD) modelling with the Random Forest (RF) algorithm to evaluate the environmental impacts of recycling concrete waste generated from building demolition, addressing key sustainability challenges. The RF algorithm validates the SD model's predictions, identifies influential parameters, and enhances predictive reliability. Sensitivity analysis examines critical variables across diverse scenarios, facilitating targeted policies to mitigate adverse environmental consequences. The RF model achieved high predictive accuracy, with R-squared values of 0.9818–0.9952 and NMSE scores of 0.0048–0.0182 for Total Landfill Consumption (TLC), Total Material Conservation (TMC), Total Primary Energy consumption (TPE), and Total CO₂-equivalent emissions (TCO2-eq).Findings reveal dual benefits of concrete recycling: conserving raw materials and reducing landfill reliance. Expanding recycling plant capacity significantly enhances recycled aggregate production and reduces landfill demand. Strategic siting of facilities near demolition sites minimizes transportation needs, supporting long-term waste reduction goals.This study highlights the critical role of scaling recycling infrastructure to accommodate increasing construction and demolition waste while reducing the environmental impacts of landfill. The integrated framework provides actionable insights for sustainable policy development, advancing understanding of the environmental impacts of concrete waste recycling.