The purpose of this research is to present a fuel calculation and estimation model for air transport operations. For this purpose, the primary data of the C-130 aircraft based on the book Dash One Performance was collected to determine the values of independent variables (weight, aircraft speed, altitude, air density, air pressure, wind speed, outside air temperature, engine air temperature) and the variable value dependent (fuel consumption) in different flight modes (landing, cruise, take-off, taxiing). Then, a model based on the MLP neural network was presented to verify and predict the fuel consumption of the aircraft. Then, flight variables were optimized using genetic algorithms. The obtained results indicate that the error value in the proposed algorithm was 0.0062. By increasing the number of hidden layers, the error value decreased to 0.0014. Also, the graph of the obtained results indicates a small difference between the results of the algorithm and the real results, which shows the performance of the algorithm. According to the results obtained from the genetic algorithm, it can be said that considering all the independent variables at the same time makes some variables ineffective in terms of fuel consumption, so their coefficients are zero. Also, the coefficients of some variables are negative, which indicates the reverse of fuel consumption and the increase of these coefficients