In concrete, the utilization of Ground Granulated Blast Furnace Slag (GGBS) as partial replacement for cement is becoming popular as it reduces the cement content used in the production of concrete and thereby reduces the cost of construction and also the carbon footprint caused by cement in concrete. GGBS, to be used as a partial replacement to cement in concrete, the strength parameters of partially replaced GGBS concrete is to be studied in details. In the present study, Multiple Linear Regression (MLR) and Artificial Neural Network (ANN), the two major statistical models are developed for the experimental values of compressive strength of concrete specimens with cement being partially replaced by GGBS up to 70%. Here, the experimentation involves both high strength (M40 Grade) and low strength (M20 Grade) concrete specimens for comparison in Artificial Marine environment and Normal Environment. In addition to all the above parameters, the compressive strength of concrete specimens subjected to Acid Attack and Sulphate Attack is also studied and included. The predicted results of compressive strength from both MLR and ANN are compared for their statistical significance and accuracy to decide the best statistical model out of both for the prediction of compressive strength of concrete.

Concrete, Multiple Linear Regression, Artificial Neural Network, Compressive Strength