STUDY ON MACHINE LEARNING METHODS FOR COMPRESSIVE STRENGTH PREDICTION USING CONCRETE MANUFACTURING CONTROL DATA

Production control data for concrete produced at a precast concrete plant were collected and machine learning was used to predict compressive strength and evaluate the results. Random Forest and Elastic-Net regression were found to be effective prediction methods, and data on specimen weight was found to be valid. It was also shown that the prediction accuracy can be improved by measuring features that were not measured in this study.