A contribuição da inteligência artificial no gerenciamento de voo

Abstract

Decreasing flight delays is to allow, among other factors, the growth of the airline sector, as with the aircraft flow going better, the entire system works more fluidly, allowing the entry of new companies in the sector and increasing its service users. However, one of the biggest bottlenecks for this growth stems from congestion in the Terminal Control Areas, known by the acronym TMA. Considering this reality, this study aims to identify the Artificial Intelligence use to reduce such congestion to the extent that it can act on aircraft, airline procedures and air traffic control. To achieve this goal, a descriptive methodology with a qualitative approach was adopted, as well a documentary and bibliographic procedures. Thus, based on the researched material, it was proven that Artificial Intelligence systems such as Garmin Autoland, ATTOL and MyFuelCoach, allow the pilot to manage both the flight and what relais on it, such as fuel consumption. In the airlines' field, intelligent systems are used mainly in aeronautical maintenance with the purpose of anticipating and preventing failures, thus avoiding delays; examples of these systems are the Prognos software and the Skywise platform. As for the use of Artificial Intelligence for air traffic management, it was found that the International Civil Aviation Organization created the Global Air Navigation Plan, which uses Performance Based Navigation, in addition to a high-level system called Total Airspace and Airport Modeler. These programs offer intelligent systems for solving congestion problems. It should be noted that in Brazil those two systems are adopted, as well as the SIRIUS implementing to reduce restrictive measures usage, such as waiting. Thus, the hypothesis that Artificial Intelligence can minimize the flight management problems in aspects related to the pilot, the airline and air traffic is confirmed; however, the field is still fertile and demands a Big Data capable of absorbing all industry information so that it can be treated in intelligent systems. Given the above, it is suggested for future studies to deepen into the evolution of a data generation and sharing systems that can further improve the Big Data and provide the evolution of Artificial Intelligence that is dependent on it and has its performance improved when fed with new information.