Flight delay prediction model

Flights in around the world have encountered with delays usually. Such delays have many costs for airlines, pas sengers and society. So identifying and predicting the delays is very important problem. Purpose of this research is to predict airline delays and identify the most important factors which generate delays in airline operation. A new data mining approach has been used for delay prediction. This approach can predict occurrence a delay and amount of each delay in the flight network. Proposed methodology was implemented on a real data set of a large Iranian airline with more than 50 airplanes that fly between 52 airports. Results show that the age of fleet, aircraft type and departure time have the most effect on delays. The accuracy of approach is 70% at predicting a delay event and predicting the amount of delay