With the growth in the Indian aviation sector, demand forecasting is a necessity for operating efficiency and strategy planning. This research concentrates on passenger and freight demand forecasting country-wise using data from 2015–2024 employing advanced regression techniques and machine learning models like ANN, LSTM, CNN, and RNN. Metrics used in evaluating the dataset include MSE, RMSE, MAE, and R-squared values. Key results indicate that for “Passenger to India,” “Passenger from India,” and “Freight to India,” the LSTM model outperformed others as it can capture temporal dependencies well, and the RMSE values are 0.4983, 0.5011, and 0.5017, respectively. For “Freight from India,” the RNN model performed the best with an RMSE of 0.5002, showing its strength in modeling sequential patterns. Although ANN and CNN models are performing competitively, the models were not so good at handling time-series dynamics compared to LSTM and RNN. The proposed study is helpful for policymakers and industry stakeholders in showing that country-wise data will be important for accurate demand forecasting. This allows better resource allocation and informed decision-making. The study has limitations, including dataset inaccuracies due to the disruptions caused by COVID-19 and the exclusion of external macroeconomic indicators. Future research can include these factors and hybrid machine learning approaches for further optimizing the predictions in order to set a benchmark for demand forecasting in the Indian aviation industry.