Predictive Maintenance IoT System for Industrial Machines using Random Forest Regressor

In this paper, we present the predictive maintenance Internet of Things (IoT) system for industrial machines. The proposed system collects electrical data including voltages, currents, and powers via LoRa nodes; and also collects specific data including manufacturers, years of manufacture, and types of machines. Then, the system makes maintenance prediction based on these data. The maintenance prediction uses the Random Forest algorithm to estimate the Remaining Useful Life (RUL) of wide-type industrial machines. Depending on the predicted values, the system will issue the warning messages for the managers to make an appropriate maintenance plan. The experimental results show that the normalized RMSE of the system can achieve up to 0.1427.