EMPOWERING THE AVIATION INDUSTRY WITH FEDERATED LEARNING FOR FLIGHT DELAY PREDICTION

Abstract

The air transport system is the most common mode of transportation among people from one country to another or from one big region to another. This enables people to move around in quick and convenient manners for both leisure and work purposes. The best performing airlines earn themselves a good reputation through effective policies, cleanliness in terms of health, participation in the community, and constant innovations in services. There are benchmarks that these airlines set in terms of service delivery and excellence in the aviation industry. Satisfaction of customers influences the policy decisions of airlines. Travel is made up of a number of elements which determine the reputation of the airlines and their success within the competitive world. Airlines try as much as possible to avoid any delay before take-off and after landing in order to ease the fears of the travelers and ensure they have a safe flight. They strive for a good travel experience using strategies and processes that would help them achieve this goal. However, despite all the operations going according to the plans, a lot of things may lead to delays. In order to measure the effect of flight delays on an industry, there is a need to collect statistics. Given the numerous sources of data, millions of pieces of data are present within the industries. The amount of data on the reasons for flight delays is so vast that human intervention is likely to result in mistakes in analysis. It requires automation of the process to analyze the data. Conventional methods of doing this will not be much helpful owing to the increased possibility of errors. Data Science and Machine Learning, the two important branches of Artificial Intelligence, help in analyzing huge amounts of data.