The demand for cement concrete usage for modern building construction is increasing gradually. The raw materials available naturally for the manufacture of cement is degrading and the main drawback in the manufacture of cement is the release of carbon dioxide gas in excess into the atmosphere in order to overcome this, an incidential product from industry act as an alternative binder namely Ground granulated blast furnace slag (GGBS), metakaolin, flyash, silica fume certainly untilized. Concrete made up of flyash or GGBS is called geopolymer concrete. These binders vary in chemical properties, which influence the compressive strength, so the forecasting is needed to determine the compressive strength. This paper’s primarily focus is to present an integrated approach for forecasting compressive strength of GGBS based sustainable geopolymer concrete by using machine techniques via Artificial Neural Network (ANN). Flyash, GGBS, Al/Bi, SS/SH, Molarity, CA, FA, SP, water are taken in variables and the compressive strength is the yield value; to train the network feed-forward and back-propagation algorithm is adopted. The developed ANN model predicts the compressive strength with a correlation coefficient values of 0.947-0.977, with the acceptable mean relative error. The average percentage errors are found to be as 0.574 % for 7 days and 0.946 % for 28 days of compressive strength by using mean absolute percentage error, root means square error. The developed model is found to be capable of forecasting the compressive strength of sustainable geopolymer concrete mixed with GGBS binder with a limited number of available data. © 2020 Alpha Publishers. All rights reserved.