Development of an ANN Model for Predicting Optimum Mix Proportion of Fiber-Reinforced Geopolymer Concrete and Assessment of the Model’s Prediction Efficiency

The environmental footprint of ordinary portland cement production is responsible for nearly 8% of global CO2 emissions, which has raised significant concerns in recent decades. In pursuit of sustainable alternatives, geopolymer concrete produced by industrial byproducts such as fly ash, ground granulated blast furnace slag, and wood ash offers an eco-friendly solution, and much of the research has already been carried out. The optimization of best mix proportion is a time-consuming process since many different binder proportions have been used by previous researchers. This study aims to develop an artificial neural network (ANN) model to optimize the mix design and predicting strength parameters. A total of 250 data sets were compiled from literature and experimental studies, considering parameters such as binder content, alkaline solution ratios, and fiber combinations. An ANN model was developed using the Levenberg–Marquardt algorithm with MATLAB to forecast compressive, flexural, and split-tensile strengths. Experimental validation was performed on five new mix combinations, and the model achieved a high correlation coefficient (R≈0.985), with 88% of predicted values showing less than 10% deviation from lab-tested results. The novelty of this work lies in integrating ANN-driven predictive modeling with hybrid fiber-reinforced geopolymer concrete incorporating wood ash, which is a relatively unutilized supplementary binder. The study not only demonstrates the robustness of ANN in predicting mechanical properties accurately but also provides a cost-effective, time-saving method for sustainable mix design. In conclusion, the proposed model has high potential for accelerating the lab-to-market transition of eco-efficient concrete technologies in structural applications.