28-day compressive strength prediction utilizing a radial basis function model incorporating meta-heuristic algorithms

The construction industry's swift manufacturing of building materials has played a pivotal role in environmental degradation worldwide. Researchers are now directing more focus towards investigating the environmental benefits of utilizing recycled concrete aggregates sourced from demolished structures. Despite notable disparities from their natural counterparts, these recycled aggregates hold promise in mitigating environmental impacts. This study is focused on a comprehensive evaluation of substituting natural fine and coarse aggregates with recycled alternatives and its direct impact on the compressive strength of concrete over a 28-day curing period. Leveraging advanced machine learning methodologies, particularly the Radial Basis Function (RBF) model, this research aims to ascertain the feasibility of integrating recycled concrete aggregates into construction practices while ensuring structural integrity. The integration of the Gold Rush Optimizer (GRO) and the Self-adaptive Bonobo Optimizer (SABO) synergistically with the RBF model enhances the precision and efficiency of the assessment process, facilitating a deeper understanding of the complex interplay between material composition and concrete strength characteristics. Through this innovative approach, the study endeavors to provide valuable insights that can inform sustainable construction practices and contribute to the optimization of resource utilization in the construction industry. According to the findings, the RBSA model, a fusion of the RBF model with SABO, showcased a remarkable correlation coefficient (R²) of 0.993, indicating an exceptionally high level of accuracy. Moreover, it demonstrated the lowest Root Mean Square Error (RMSE) value, registering a mere 1.244. These results underscore the outstanding predictive prowess and optimization of the RBSA model in forecasting f’c-28, affirming its reliability and efficacy.