Research on Flight Fuel Prediction based on Historical Data Mining

With the rapid development of the air transport industry, aircraft fuel consumption continues to increase. Reducing unnecessary fuel consumption by more accurately loading aircrafts with fuel is emerging as an important consideration to enable airlines to optimize cost management and measures in response to energy conservation and emission reductions. This paper uses predominantly historical flight data to forecast fuel requirements for flights concerning a period of time in the future. Firstly, the historical data of the flight is analyzed and several prediction methods that are applied to the data mining are summarized. Combining the characteristics of the historical data of the flight and the actual fuel forecast, multiple linear regression is used, and a new fuel forecasting model is established. This proposed model could provide a technical reference for airline fuel management strategy and flight plan formulation in the future.