Machine learning approach for predicting the influence of pulverized waste tire rubber fines as a sustainable cement alternative in concrete.

The disposal of waste tires has a significant environmental impact, and there is growing interest in their integration as supplementary cementitious materials. Using recycled waste tire fines as a cementitious material can declare that rubber addition develops a practical pathway for valorising tire waste in concrete production without compromising the strength of concrete. In this study, pulverized waste tire rubber fines were incorporated as a substitute for cement at varying dosages, and their influence on the mechanical behavior of concrete was investigated. The microstructural arrangements of the waste tire concrete were examined to ensure the appropriate reactions between waste tire rubber and the rest of the materials and gelation matrix development. To compute the influence of waste tire rubber fines on the observed nonlinear pattern of experimental results, a transformed square root statistical approach was applied for the significant predictions and effect correlations. From this approach, predictive transform models C1 and F2 achieved better predictions with a higher significance level. The key findings are optimal rubber dosage for maintaining satisfactory mechanical properties and instigating an effective waste disposal pathway. The machine learning model results were in good association with nonlinear experimental trends and highlight the beneficial replacement range of waste tire rubber fines.