Accurate forecasting of air passenger traffic is crucial for the efficient management of airline operations and airport infrastructure. This paper explores the application of autoregressive models to predict air passenger volumes. Leveraging historical data, we analyze the performance of AR and autoregressive integrated moving average (ARIMA) models. Our results on the aviation dataset demonstrate that these models effectively capture the underlying patterns and seasonal fluctuations in passenger traffic, providing reliable short-term and long-term forecasts. Through comprehensive evaluation metrics and comparative analysis, the study underscores the potential of autoregressive models as robust tools for air passenger forecasting, aiding stakeholders in making informed decisions and optimizing resource allocation.