Compressive strength prediction of fly ash-based geopolymer concrete via advanced machine learning techniques

Concrete is a widely used construction material, and cement is its main constituent. Production and utilization of cement severely affect the environment due to the emission of various gases. The application of geopolymer concrete plays a vital role in reducing this flaw. This study used supervised machine learning algorithms, decision tree (DT), bagging regressor (BR), and AdaBoost regressor (AR) to estimate the compressive strength of fly ash-based geopolymer concrete. The coefficient of determination (R2), mean absolute error, mean square error, and root mean square error were used to evaluate the model's performance. The model's performance was further confirmed using the k-fold cross-validation technique. Compared to the DT and AR model, the bagging model was more effective in predicting results, with an R2 value of 0.97. The lesser values of the errors (MAE, MSE, RMSE) and higher values of the R2 were the clear indications of the better performance of the model. Additionally, a sensitivity analysis was conducted to ascertain the degree of contribution of each parameter towards the prediction of the results. The application of machine learning techniques to predict concrete's mechanical properties will benefit the area of civil engineering by saving time, effort, and resources.