In this project, a machine learning model is built to forecast one-day ahead hourly power demand of the Peruvian national electric system (SEIN) using time series historical data. By predicting power demand, free users (large consumers) can make better decisions such as optimizing their “coincident peak demand” (between 5:00 pm and 11:00 pm) at the time of the monthly national system's peak demand and, therefore, reduce their transmission charges. Based on the characteristics of this forecasting problem, the performance of the model will be measure using the RMSE metric.