PREDICTION OF THE SLUMP AND STRENGTH OF HIGH STRENGTH CONCRETE USING RANDOM FOREST MODEL

Abstract

Random Forest (RF) has been successfully applied to a variety of engineering problems due to its simplicity, versatility, and suitability for both classification and regression tasks. Concrete, as a material composed of multiple complex elements, is influenced by numerous factors, posing challenges in accurately predicting its properties. In this article, an RF model is developed in predicting the slump and strength of concrete using mixed mineral admixtures from blast furnace slag and silicafume. The criterions to evaluate the accuracy of the models are the R squared (R2) and the root mean square error (RMSE). Comparing the predicted data with the tested data, the result indicates that RF model should be used in predicting both the slump and strength of concrete.