Basalt fiber-reinforced concrete ( \(BFRC\) ) provided a type of high-performance concrete. In order to address the limited availability of research on the flexural strength ( \(FS\) ) of \(BFRC\) , it is necessary to create and evaluate approaches for predicting \(FS\) . The current research examined three approaches, including Support Vector Regression ( \(SVR\) ), Random Forests ( \(RF\) ), and Multivariate Adaptive Regression Spline ( \(MARS\) ), for the purpose of assessing goals. Due to the fact that the Chimp algorithm ( \(CA\) ) is related with algorithms in order to discover the most efficient combination of hyperparameters, the accuracy of this simulation is largely dependent on its hyperparameters. Future studies should gather a more diversified data collection of \(BFRC\) formulations, curing processes, environmental variables, and fiber qualities. To find the best machine learning method for \(BFRC\) ’s \(FS\) prediction, future research should go beyond applied algorithms. The results showed that there is great potential for reliably forecasting the \(FS\) of \(BFRC\) using the hybrid and optimized approaches. According to that setup, \(RFCA\) obtained \({R}^{2}\) values that were higher—0.983—than \(SVRCA\) and \(MARSCA\) during both the learning and assessment phases. Significant gains were shown in the comparison of the findings from this study and \(LightGB\) (literature). Overall, the findings and justifications prove that the developed analysis could predict precisely the target with the superiority of \(RFCA\) , where can be applied for practical applications.
