Laboratory-on-IoT for predicting construction temperature of asphalt pavement

Construction temperature of asphalt pavement is an important factor affecting the quality of construction. The gradually mature Internet of Things technology and machine learning method provide new support for improving the quality management of asphalt pavement construction. Based on the pavement construction data collected by the Internet of things, the importance assessment shows that the discharge temperature and the speed of paving and rolling significantly impact the construction temperature. In addition, the model framework and hyperparameters optimization of the neural network prediction model are carried out. The results show that the Adam optimizer is superior to the other three optimizers. The addition of batch standardization helps to improve the model training efficiency, and the optimization of hyperparameters significantly improves the model performance. The combined system of Internet of Things and machine learning can be used for real-time prediction of construction temperature and can provide guidance for construction temperature regulation. • Machine learning can predict the construction temperature of asphalt pavement. • The importance of features is evaluated based on random forest. • Framework and hyperparameters of the model are optimized.