I am curious about how the arrival of flights would impact the pushback time for departure flights. As far as I know, each flight has a unique identifier known as GUFI, and it is not possible to determine how to link an airplane from an arriving flight to a departing flight. Despite the abundance of data for arrival flights, I feel like I am missing something important. For instance, there are numerous estimations in TFM and TBFM for arrival flights, and the only purpose they seem to serve is to determine how busy a particular airport is. Furthermore, there is a first_position table for arrival flights, and the arrival_stand_actual_time field in the _standtimes table indicates the time when a flight arrived at the gate in the destination airport. Based on the training data provided by the competition, it appears that there are more data available for arrival flights than for departures. Is there something I am missing, or could you offer some insight/intuition on how all this data for arrival flights are relevant to the problem at hand?
As far as I know, each flight has a unique identifier known as GUFI, and it is not possible to determine how to link an airplane from an arriving flight to a departing flight.
Correct that that information is not directly provided in the competition data.
Is there something I am missing, or could you offer some insight/intuition on how all this data for arrival flights are relevant to the problem at hand?
I don’t think I have anything particularly insightful to add, just that you’ve pointed out a central question of the challenge: what are the patterns of activity at an airport and how can we identify patterns that result in delays to pushback time? Arrivals are definitely one piece, and you might be right that the best use for arrival data is to get a sense for “how busy” and airport is. The question of “how busy” alone certainly can be addressed in many creative ways. There might be other ways to make use of the arrival data as well. Sorry that I don’t have anything more profound than that to say, but I don’t think you’re missing anything, and filling in the blanks is all part of the challenge!