In an effort to manage and optimize the use of energy in these buildings, the ARIMA model (Autoregressive Integrated Moving Average) emerged as a very important analytical tool. The primary objective of this research is to investigate and explore how ARIMA parameter settings can be modified to improve the accuracy of energy consumption predictions on stairwell buildings. Based on an in-depth analysis of existing literature as well as empirical data, it was found that the ARIMA model, by leveraging time row kastasioneran and the use of univariate data, showed great potential in producing highly accurate short-term predictions. In this study, it was found that by performing the correct configuration of ARIMA parameters, the model was able to a level of accuracy with MAPE of 5,317% and RMSE of 8,7. These results show an excellent level of conformity, indicating that the ARIMA model can be effectively used to improve the accuracy of prediction of energy consumption in stairwell buildings. The findings of this study confirm that with proper adjustment of parameters, ARIMA can be a very useful tool in more efficient energy management in the building sector, which can ultimately contribute to reducing unnecessary energy consumption as well as improving overall energy efficiency.