In this paper, A hybrid SARIMA-RF model is proposed to the practical problem of the air travel demand forecast, considering the scenario of carrying out new route. The Seasonal Auto Regressive Integrated Moving Average (SARIMA) model is generated to deal with the air travel demand, while the Random Forecast (RF) is generated as the demand stimulus model to deal with those markets with the influences of new routes, which the SARIMA can not give a good forecast result. The experiment with actual airline market data is carried out to validate the performance of the proposed model, and the results show that the SARIMA-RF model has a better performance with smaller forecast error than the original SARIMA time series forecast model in the markets with the new routes.