GE crowdsources air traffic data analysis with Flight Quest 2
GE has just revealed the private leaderboard on Flight Quest 2 -- and four outstanding winners from the Kaggle community. Flight Quest 2 began last August as the second part in GE’s Industrial Internet Flight Quest, in partnership with Alaska Airlines and Kaggle. They challenged data scientists around the world to develop algorithms to increase flight efficiencies in real time, reducing delays and maximizing a flight’s profitability. Using national airspace data provided by Flight Stats, the winning Flight Quest 2 algorithms determine the most efficient flight routes, speeds and altitudes at any moment taking into account variables such as weather, wind, and airspace restraints. The winning model by José Fonollosa proved to be up to 12 percent more efficient when compared to data sets from actual flights.
GE plans to develop rapid prototyping from these algorithms to deliver a first-of-a-kind solution that will give pilots and dispatchers a prescriptive, network-wide view of flight route variables, which does not exist today, and will allow airlines to recognize and adjust flight routes in real time. GE determined that if each scheduled flight worldwide reduced the distance it flew by only 10 miles, airlines could reduce annual fuel consumption by 360 million gallons and save the industry more than $3 billion each year.
Through Flight Quest, GE tapped Kaggle’s worldwide community of data scientists and received more than 6,800 combined submissions across 58 countries, with a total prize pool of $500,000 from GE.