A statistical study of the weather impact on punctuality at Frankfurt Airport.

A hybrid regressiontime series modelling was used to relate the total daily punctuality TOTP at Frankfurt Airport, Germany, to weather, the traffic flow and the airport system state. The selected modelling approach is applied to the annual, the multiannual and seasonal data of the years 2001–2006. Reduction of the initial variables system dimension is achieved through the application of the principal component analysis PCA method. The portion of the variability that can be explained by the model after correction of autocorrelations in the residuals using autoregressive AR models, is between 60 and 69. A slight yeartoyear variability in the number of statistically significant model parameters is detected. Upon identification of the terminalspecific delay impact factors, the possibility of the 24 h punctuality forecast for the interval of 1 year was discussed. Copyright © 2008 Royal Meteorological Society [ABSTRACT FROM AUTHOR]