SlideShare utilise les cookies pour améliorer les fonctionnalités et les performances, et également pour vous montrer des publicités pertinentes. Si vous continuez à naviguer sur ce site, vous acceptez l’utilisation de cookies. Consultez nos Conditions d’utilisation et notre Politique de confidentialité.
SlideShare utilise les cookies pour améliorer les fonctionnalités et les performances, et également pour vous montrer des publicités pertinentes. Si vous continuez à naviguer sur ce site, vous acceptez l’utilisation de cookies. Consultez notre Politique de confidentialité et nos Conditions d’utilisation pour en savoir plus.
Airport Forecasting - 2020
(Issue No. 45)
Toulouse Airport (TLS)
There is no means to forecast, if we do not hold it in the right time.
Long time ago as we were in the schools we learnt that – 80% value of R – square was accepted level
for best fitting for the Data. Today, in practice and real life, the margin of fitting is too tight, and the
change in values of R- square has a significant meaning on final results. Therefore, we are exploring
all the possible option of the outcomes – that put the top management on a solid ground to stand and
get a clear picture for the future.
Annual Passengers Forecast:
There are three possible outputs for forecasting, positive growth, leveling (Zero Growth) and negative
trend, and the best way to set up annual target and minimize the data discrepancy is to address the
data by two trend models using the concept of 12 rolling months.
First – General Trend Model using the concept of Straight Line equation – defining general trend.
Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation.
This reflects the impact of most recent data on the path of general trend. The mid-point is the most
convenient forecast annual result at Dec 2020. So as long as the gap between two models is small, the
more accurate approaching value for setting annual target otherwise we have to select the half way
distance between two extreme targets of these two models provided that Dec 2020 > Dec 2019.
Scenario 1: Preset Annual Target = Passengers = 9,673,392 Pax.
At annual growth 1.2 % and R- square = 95.01 % and boundary error range -5.18 to +6.57 %, we
select preset value for Dec 2019 = 9,673,392 as Dec 2019 > Dec 2020 for first scenario. –
Scenario 2: Optimum Solution. = Passengers = 9,969,622 Pax.
This is an optimum solution without any pre-set targets or constrains that governed the analysis, the
outcome results = Pax 2020 = 9,969,622, at annual growth 2.42 %, R- square = 95.56 % and
boundary error range -3.01 to +5.57 %.
(below : how to read the Graph)
By: Mohammed Salem Awad