Concrete mix design modelling based on variation of hidden layer and neuron of ANN for virtual learning development

Artificial Neural Network (ANN) is a Machine Learning (ML) algorithm which learn by itself and organize its thinking to solve problems. Although the learning process involves many hidden layers (Deep Learning) this algorithm still has weaknesses when faced with a high noise data. Concrete mixture design data has a high enough noise caused by many unidentified / unmeasurable aspects. Information needs about the compressive strength of early age concrete (under 28 days) are often needed while the construction process is still ongoing. H2O’s Deep Learning type of ANN has been tried to predict the compressive strength of early age concrete, but the results are less than optimal. This study aims to improve the H2O’s-ANN prediction model using Bagging to reduce the influence of noise and overfitting. The lowest RMSE that are able to be achieved in this research with Bagging is 6.385 while it is 6,674 without Bagging. This result proves that Bagging is significant to reduce the Deep Learning error rate for predicting the compressive strength of early age concrete. Future work this model, as a new innovation in machine learning in civil engineering vocational education subject, can be used to various concrete mixture design and learning about concrete. Because this concrete mix design model is digital, it can be used for virtual learning. Currently the author is conducting research to create a virtual concrete mix design learning model.