Predicting the Strength Properties of LWC Using Response Surface

The current research uses the response surface method to estimate the mechanical properties of lightweight concrete with various admixtures (i.e., silica fume and metakaolin). The mechanical characteristics of lightweight concrete were computed by considering the different input variables: (i) metakaolin, (ii) silica fume, and (iii) curing age. RSM models were built on the experimental outcomes of LWC mixtures at 28 days of curing age. The reliability and prediction accuracy of the proposed probabilistic models are examined using the experimental data. It is demonstrated that using these new RSM is just as easy as using any fundamental formulae, but they offer a better tool for predicting strength properties. Regression models were tested for responses using the residual plots and main effects. Each progression variable’s statistical significance was assessed, and the resulting models were articulated in a second-order polynomial equation. Two findings demonstrated that including all admixtures improved the strength qualities, while higher levels of incorporation reduced the strength. The error percentages of the validation tests are <4% and 3% for compressive strength and split tensile strength, respectively.