Experimental and analytical optimization of geopolymer bricks incorporating GGBS

Purpose Construction waste disposal is becoming an increasingly important aspect of civil engineering. Waste disposal in the construction sector comes from demolition sites and research projects. To some extent, predictive models are advised to reduce waste in research projects. Design/methodology/approach The study aims to provide a novel prediction approach the Spider LeNet Predictive System (SLPS), for compressive strength, apparent density and water absorption in geopolymer brick manufactured from fly ash, sand, alkali-activator and various quantities of Ground Granulated Slag generated from Blast Furnace (GGBS). The amount of GGBS varies with the fraction of fly ash (0%, 5%, 10%, 15% and 20%), and the brick samples are cured at ambient temperatures of 60°C, 80°C and 120°C. Initially, experiments are conducted. The experimental results are incorporated into the MATLAB software, and prediction models are developed thereafter. The prediction models are then applied to the innovative SLPS to predict compression strength, bulk density and water absorption in geopolymer bricks. Findings The brick composition of 50% fly ash, 20% GGBS and 30% sand is anticipated to have a compressive strength of 31.94 MPa after 28 days of curing at 120°C, with water absorption of 2.05%. The highest bulk density in the geopolymer brick was anticipated to be 1.838 g/cm3 for the seven-day specimen. Originality/value The results show that the suggested technique accurately predicts the properties of the geopolymer block.