Sand Cat Swarm Optimization-based prediction regressions on the elastic modulus of Recycled brick aggregates concrete

Recycled brick aggregates (RBAs) are produced by the process of crushing unused bricks.This approach offers an effective solution for addressing environmental degradation and limited availability of natural resources in civil engineering sector. The objective of this study is to promote the widespread adoption of recycled brick aggregate concrete (RBAC) in the construction industry. The elastic modulus (E_RBA) of concrete produced from recycled brick aggregate is calculated by applying three alternative methodologies. This research presents a comparison of the Sand cat swarm optimization (SCSO), a hybrid optimal approach when utilized for two standard machine learning methods: Random Forests Analysis (RFA) and Gradient Boosting Regression (GBR). The determination of the elastic modulus of recycled brick aggregate in concrete involves the use of six distinct factors. The variables are determined using a computerized database that contains 123 test results from prior research.This approach was implemented to ensure a thorough evaluation of the framework. Based on the presented metrics in the training section, PI and U (95%) decreased from 0.0345 and 4.5938 related to RFA(S) to 0.0385 and 5.1174 related to GBR(S), respectively. According to the results, the outcome of metrics (PI and U_(95%)) in the previous sentence, for the testing section, it was reduced from 0.0301 and 3.4011 related to RFA(S) to 0.0338 and 3.7992 related to GBR(S). By comparing all the indicators and total of scores for metrics in two models called GBR(S) and RFA(S), it is evident that model RFA(S) outperformed model GBR(S) (total scores for metrics in GBR(S) and RFA(S) are 16 and 32)