Geopolymer concrete (GPC) could be used as an environmental-friendly alternative solution for concrete production due to the detrimental impacts of cement production on the environment. Since obtaining an optimal mix design, and subsequently, predicting the compressive strength of mortar using experimental means is costly and time-consuming, employing soft-computing techniques could facilitate and accelerate this approach. In this research, gene expression programming (GEP) was used to develop numerical models for predicting the compressive strength of GPC based on ground granulated blast-furnace slag (GGBS). Through an experimental program, an extensive database of the compressive strength of GGBS-based GPC consisting of 351 specimens obtained from 117 different mixtures was generated. The five most effective parameters, including specimen age, sodium hydroxide (NaOH) solution concentration, natural zeolite (NZ), silica fume (SF), and GGBS content, were considered as the modeling input parameters. Using GEP algorithm, simplified and practical mathematical equations were proposed to predict the compressive strength of GGBS-based GPC mortar. Performance, high accuracy, and predictability of the proposed equations were validated by the conducted sensitivity and parametric analyses. The obtained results could promote the re-use of GGBS for GPC development, which in turn will lead to environmental and economic advantages.
