Utilization of a Bayesian Network Algorithm to Predict the Compressive Strength of Concrete

Artificial Intelligence (AI) technology, includingArtificial Neural Network (ANN), Fuzzy Logic (FL), and Random Forest (RF), have been utilized widely in past literature in predicting the behavior of concrete products.Few studies used the probabilistic inference approach through Bayesian Networks (BN) to envisage the structural health integrity of concrete, while the possibility of employing BN algorithms for the prediction of its mechanical properties has not been investigated yet.This research evaluates the potential applicability of BN in predicting the compressive strength of selfcompacting concrete made with various supplementary cementitious materials and basalt fibers.Two learning algorithms, namely Naïve Bayes and Markov Blanket, were employed along with various discretization methods to maximize network performance and minimize integral absolute error.Research findings showed that the Naïve Bayes classifier coupled with K-means discretization tool with 4 segments of 'days' data and 3 segments of the remaining variables gave the highest correlation between experimental and predicted values.Meanwhile, the Markov Blanket algorithm failed to accurately predict the compressive strength.The accuracy of the predicted BN was found to be comparable to that obtained from an ANN model.