Delay
Forecast 2004, based on national capacity
enhancement plans
The FAP
(Future ATM Profile) model was used to forecast the 2004 Summer Delay in each
ECAC (European Civil Aviation Conference) centre.
The input data
were:
·
the
STATFOR (Eurocontrol
Statistics and Forecasts Service) traffic forecast and
·
area
control centre (ACC) capacities as committed by Air Navigation Service Providers
(ANSPs).
The result was a prediction of further
reduction in delay, especially in en-route delay per flight, from 1.8 minute
per flight in 2003 to 1.5 minute in 2004 (0,8 min. en-route + 0.7 airport).
An EEC note will be issued in April.
Delay forecast study
Since 1999,
the FAP (Future ATM Profile) model has provided the EAG (European ATFM
Consultation Group) each spring, with a delay forecast for the following
summer.
FAP uses CASA
(Computer Aided Slot Allocation), the operational CFMU tool, to compute delay
based on traffic and capacities. The originality of the FAP model is to work at
macroscopic level i.e. at ACC level and to take into account the "network
effect" (the mutual interaction between centres), which is particularly
important in a delay forecast.
Therefore, the
two main inputs are:
·
the
forecasted traffic, derived from the STATFOR prediction (+3.9% in the baseline
scenario for the ECAC area) and
·
the
ACCs planned capacities as declared by the ANSPs in their LCIPS (Local
Convergence and Implementation Plans).
The results
are provided at ECAC level and also for each individual ACC. The study can be
summarised as follows:
Outcome
After 5 years
of utilisation, it is possible to get an idea of the validity of this
prediction. The figure below shows the comparison between forecasted and observed
delay:
The
coefficient of correlation between forecasted and observed delays is 0.82. It
might be said that FAP gave a good indication of the trend, even if it must be
recognised that some effects might have compensated one another (in general, lower
traffic than predicted, especially in 2001, but also lower capacity increases
than committed).
Anyway this
delay-forecast study is not made for the sake of forecasting. Its main interest
was to show the high elasticity between capacity and delay: in certain centres
1% lack of capacity can cause up to 25% extra delay. These results encouraged
some ANSPs to make, during the spring, additional short-term effort to avoid
huge delay.
This might
have contributed to the significant reduction in delay per flight, as it can be
seen on the previous figure, during the last 5 years.
For further
information contact Marc Dalichampt