Improving schedule reliability based on copulas: An application to five of the most congested US airports

This study, based Newark Liberty International, New York John F. Kennedy, New York LaGuardia, Chicago O’Hare International and San Francisco International in the US, provides a perspective on the predictability of on-time performance evaluation by applying an actuarial and financial method called copula. It is common for the percentage of on-time gate arrivals to be portrayed in daily news as an indicator of schedule reliability. However, the degree of interdependency between gate arrival delays and block delays may provide some indication of schedule reliability.

► Since two variables’ distribution is not normal, correlation statistics cannot measure their linear dependence.
► Concordance between two non-normally distributed variables can be measured by copulas.
► The degree of concordance between block delays and gate arrival delays serves as a measure of flight schedule reliability.
► The research is based on a sample of the five most delayed airports in summer 2009 in the United States.