Predicting of geopolymer concrete compressive strength using multiple linear regression method

Abstract

Concrete is a construction material that has uncertain quality and must be designed with such a composition to achieve the expected quality. Geopolymer concrete is an environmentally friendly concrete obtained by replacing Portland cement with fly ash as a binder. Geopolymer concrete requires an alkaline solution as an activator in the polymerization of aluminum and silica. The determining process of the concrete mix composition is called mix design. Mix design for normal concrete is regulated in SNI 7656:2012 while for geopolymer concrete there is no specific standard that regulates it. With the development of technology and information, existing geopolymer concrete data can be used for mix design modeling. With the data of the geopolymer concrete mixture processed using the SPSS multiple linear regression method, the regression model obtained is Y = 0.165x1 + 0.055x2 + 0.037x3 - 0.053x4 + 0.263x5 - 0.288x6 - 137.18. This regression model states that the variables x4 (NaOH) and x6 (water) have a negative effect on the compressive strength of concrete. The effect of independent variables on the compressive strength of concrete simultaneously is 29.6% while the remaining 70.1% is influenced by other factors with the standard error of the estimate value of the model is 9,60179.