This study introduces an advanced approach to time series modeling through the application of Subset ARMA (AutoRegressive Moving Average) identification. Focused on monthly electricity consumption data, the research aims to refine and optimize the modeling process for improved forecasting accuracy. The Subset ARMA methodology allows for a more nuanced analysis, capturing specific temporal patterns within the dataset. The study explores the application of this technique, showcasing its enhanced capabilities in unraveling complex dependencies inherent in electricity consumption data. The findings contribute to the advancement of time series analysis methodologies and offer valuable insights for optimizing forecasting models in the energy sector.