Estimation of 28-Day Compressive Strength of Self-Compacting Concrete Using Multi Expression Programming (MEP): An Artificial Intelligence Approach †

Abstract

Self-compacting concrete (SCC) is an innovative building material having special properties such as increased flowability, good segregation resistance and compaction without vibration, etc. Despite the benefits of SCC over conventional concrete, there are very few methods reported in the literature that can predict the 28-day compressive strength of SCC accurately. Thus, to promote the use of SCC in the construction industry, an innovative machine learning technique named multi-expression programming (MEP) was employed to forecast the 28-day compressive strength of SCC. A database consisting of 216 points was constructed using an extensive literature search. The resulting equation obtained by employing the MEP algorithm relates the compressive strength of SCC with the six most influential parameters, i.e., water–cement ratio, fly ash and silica fume, quantities of fine and coarse aggregate and superplasticizer dosage. The database was split into training and validation datasets used for the training and validation of the algorithm, respectively. The accuracy of the algorithm was verified by using three statistical error metrics: mean absolute error (MAE), root mean square error (RMSE), and coefficient of correlation (R). The results revealed that the errors were within the prescribed limits for both the training and validation sets and that the developed equation has excellent generalization capacity. This was also verified from the scatter plots of the training and validation datasets. Thus, the developed equation can be used practically to forecast the strength of SCC containing fly ash and silica fume.