The aim of this research is to design and implement a smart electricity metering system for households by utilizing the Internet of Things (IoT) and incorporating a billing prediction feature using the Random Forest algorithm. Given the increasing electricity consumption in Indonesia, developing a solution that enables consumers to efficiently monitor and manage their energy usage has become crucial. The proposed system facilitates real-time electricity consumption tracking through the Blynk application, while also providing cost estimates in Indonesian Rupiah based on recorded usage data. To provide accurate forecasts of electricity bills over a specified period, the Random Forest algorithm analyzes measurement data. Based on the results of this study, the designed smart meter system functions effectively and meets the original research objectives. The system is capable of delivering real-time electricity consumption data in monetary terms (Rupiah), along with predictions of future bills for a predefined timeframe. It can also report electricity usage data from three selected main rooms. The system achieved a Root Mean Squared Error (RMSE) of 0.12% and an accuracy rate of 99.88%. The Blynk IoT platform, used to provide real-time electricity consumption data, bill predictions, and consumption data for the selected rooms, performed excellently in line with the goals of this research..
Keywords: Smart Meter, Internet of Things, Billing Prediction, Random Forest