Fuel consumption constitutes a critical element in aviation operations, accounting for nearly 40% of operational costs. Recently, airlines have been employing a variety of methods to optimize its use. This study illustrates predictive models of fuel consumption based on time series, using flight data provided by an airline for the ATR42 aircraft fleet. These data were collected through the Post-Flight Information System (PFIS). The initial stage of the research involved a stationarity analysis on the data, establishing the need to differentiate the series. Once the series was adjusted, Autoregressive Integrated Moving Averages (ARIMA), Seasonal Autoregressive Integrated Moving Average (SARIMA), and Seasonal Auto-Regressive Integrated Moving Average with exogenous factors (SARIMAX) models were implemented, following the Box-Jenkins methodology. The effectiveness of these models was evaluated using three comparative metrics: Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Akaike Information Criterion (AIC). The interpretation of the results indicates that the SARIMAX model provides the most accurate prediction of fuel consumption data. This finding holds significant importance, as identifying an effective model for predicting fuel consumption can allow for more efficient management of operational costs in the aviation industry.