Estimation of the Elastic Modulus of Concrete Made with Recycled Brick Aggregate by Tree Algorithms

Crushing discarded bricks into recycled brick aggregates (RBAs) aids sustainability in construction. This study estimates the elastic modulus of recycled brick aggregate concrete (RBAC). In this study, a comparison is made between two hybrid optimal approaches, namely the Prairie dog algorithm (PDA) and the Brown bear algorithm (BBA), when used in conjunction with a conventional machine learning (ML) methodology known as the Decision tree (DT). Six different factors are used to calculate the elastic modulus of RBA found in concrete. The identification of these factors is accomplished by the use of a computational database that includes 123 outcomes of tests from earlier research. More specifically, the process of constructing and evaluating the suggested framework was comprised of using 75% of the data as a learning subset and a balance of 25% as a validation subset. This was done in order to ensure that the framework was thoroughly evaluated. Combining DT methods with PDA and BBA procedures (referred to as DTPDA and DTBBA, respectively) simplified the process of determining the elastic modulus of RBA, or E_RBA. It is quite likely that the E_RBA can be computed accurately using both DTPDA and DTBBA methods. R2 values of 0.943 and 0.9467, respectively, indicate that the DTPDA approach exhibited a high level of operational dependability throughout the learning and evaluation phases. The data shows that the DTBBA approach performed better than the DTPDA strategy in terms of R2 values. Nearly equal results of 0.9705 and 0.9734 were obtained using the DTBBA technique.