In this article, the demand for air travel is analyzed. By using a model, the impact of geographical, socioeconomic, and competitive aspects has been investigated while studying passenger travel demand. To this end, departure data on at Kerman airport have been gathered during the period of 2011 to 2020. First, the demand is forecasted by using an economic model. In this model the importance of significant differences of all variables is examined. Therefore, an entirely new set of data is produced and the minor variables have been removed. The K-Means clustering algorithm is then used to analyze this data, after which it is used as training data for neural network learning. The neural network used for this analysis is an LSTM Deep Learning Network, which has been used to forecast passenger demand for future coming years. Finally, with economic and social variables including GDP, income, population, inflation, exchange rate, gasoline prices and oil prices for the coming years, the percentage change in the number of passengers for each year compared to the previous year has been predicted. Based on the outputs of the neural network, changes in air travel demand are determined based on the variables of gross national product, mean income, gasoline price, and oil price inflation for each specific time. Among all these variables, the most important variable is GDP, which has a significant influence on air travel demand. The accuracy obtained in this method is 83%, which is a very good accuracy level for air travel demand compared to other regression methods.