Publication

Abstract : The accurate anticipation of the aircraft’s range is crucially important. Considering the rapid growth of the civil aviation industry, necessary steps are being taken to provide optimized performance of the aircraft as well as ensuring the safety and security of the passengers The main causes include the overall efficiency of the airplane, flight planning, and fuel management. Range prediction models based on traditional methods use complex mathematical and physics models which are often based on static parameters and simplified assumptions which lead to many possible errors, especially in the case of dynamic flight conditions. This framework suggests that machine learning (ML) algorithms can used for optimizing aircraft range prediction and provide a large number of advantages over the conventional methods. The research talks about the different ML algorithm types being utilized and what steps were taken to improve a model’s performance. The analysis of each model is made by taking various parameters into consideration. The parameters include Range [km], Fuel Non-Cr., Fuel Res. [kg], Breguet [km], Fuel/Range/Pax, Landing Mass Take-Off Mass [kg], Fuel [kg], Fuel Overhead [kg], Payload [kg] Pax, Fuel/Payload [kg/kg] to determine the Predictive Range of the Aircraft. Comparative analysis has been carried out to identify the most outstanding ML algorithm as a potential best method of predicting the range of an aircraft by means of a dataset used in this work.