As the size of the population increases, technological advancement and improved lifestyles accompany this growth as well, resulting in higher electricity consumption. There is a need to forecast the future demand for electricity in order to prepare electricity companies for their operations and provide sufficient energy for their consumers. In this paper, Multi-layered Neural Network (MLNN) and Feed-Forward Neural Network were utilized to formulate a prediction model of the monthly electricity demand of Cagayan de Oro City. The study builds a 108-months data collection from 2012-2020 as input and output data for the forecasting model. The training data set is from 2012-2019 and the testing data set is for the year 2020. The inputs for the models are gross domestic product, number of consumers, the effective rate of consumers, average monthly rainfall, and average monthly temperature. The results showed that FFNN has better performance for forecasting the monthly electric demand as indicated from the computed mean absolute percentage error. Cagayan Electric Power and Light Company, Inc. (CEPALCO) will benefit from this research by being able to supply enough power to its customers without worrying about running out.