A Multiple Linear Regression Prediction of Concrete Compressive Strength Based on Physical Properties of Electric Arc Furnace Oxidizing Slag

Various physical properties in electric arc furnace (EAF) slag will result in different concrete physical characteristics. If concrete compressive strength could be predicted using physical properties in EAF slag, both cost and time could be saved and better compressive strength could also be achieved. The better compressive strength could also be achieved. Since there was no previous study in which the compressive strength was predicted, the multiple linear regression (MLR) method was employed to predict concrete compressive strength of EAF slag in this study. When constructing the model, the minimum mean absolute percentage error (MAPE) of 3.77 % and minimum mean square error (MSE) of 4.00 could be achieved using MLR. Using MLR, it is predicted that the minimum MAPE of 2.20 % and minimum MSE of 46.95 could be achieved. Therefore, MLR could be applied successfully in predicting compressive strength. Theresults also indicated that the compositions of slag could be applied on prediction of compressive strength well.