Federal Aviation Administration (FAA) Air Traffic Flow Management (TFM) decision-making is based primarily on a comparison of predicted traffic demand and capacity (usually called Monitor/Alert Parameter, or MAP) at various elements of National Airspace System (NAS) such as airports, fixes and sectors to identify potential congestion. The current FAA Traffic Flow Management System (TFMS) and its decisionsupport tools operate with deterministic predictions and do not consider the stochastic nature of the predictions. Sector demand predictions appear to be less accurate and stable than predictions for airports and fixes. The major reason is that, unlike airports and fixes where flights are aggregated in 15-minute intervals, TFMS predicts sector demand by aggregating flights for each minute and using the one-minute peak demand as a measure for sector demand for entire 15-minute interval. This paper presents a novel analytical approach to and techniques for translating characteristics of uncertainty in predicting sector entry times and times in sector for individual flights into characteristics of uncertainty in predicting oneminute sector demand counts. The paper shows that expected one-minute sector demand predictions are determined by a probabilistically weighted average of one-minute sector entry demand predictions for several consecutive one-minute intervals within a sliding time window. The width of the window is determined depending on probability distributions of errors in flights’ sector entry time predictions. Expected one-minute sector demands along with standard deviations of demand counts are expressed via probabilistic averaging of series of one-minute deterministic predictions of number of flights entering a sector. The results of the paper contribute to probabilistic predictions of congestion in airspace. These analytical results can also be used to evaluate the impact of improved accuracy in flight timing predictions on reducing uncertainty in traffic demand predictions, hence leading to better identification of congestion in airspace.
[1]
M. Ball,et al.
Estimating Flight Departure Delay Distributions—A Statistical Approach With Long-Term Trend and Short-Term Pattern
,
2008
.
[2]
Scott Smith,et al.
Characterization of Uncertainty in ETMS Flight Events Predictions and its Effect on Traffic Demand Predictions
,
2008
.
[3]
G. Hunter,et al.
A Real-Time Probabilistic Traffic Flow Management Evaluation Tool
,
2006,
2006 ieee/aiaa 25TH Digital Avionics Systems Conference.
[4]
Anthony J. Masalonis,et al.
MEASURING UNCERTAINTY IN AIRSPACE DEMAND PREDICTIONS FOR TRAFFIC FLOW MANAGEMENT APPLICATIONS
,
2003
.
[5]
Scott Smith,et al.
Probabilistic Prediction of Aggregate Traffic Demand Using Uncertainty in Individual Flight Predictions
,
2009
.
[6]
Craig Wanke,et al.
PROBABILISTIC CONGESTION MANAGEMENT
,
2005
.
[7]
Scott Smith,et al.
Analysis of Uncertainty in ETMS Aggregate Demand Predictions
,
2005
.
[8]
Scott Smith,et al.
Probabilistic Predictions of Traffic Demand for En Route Sectors Based on Individual Flight Data
,
2010
.
[9]
Larry A. Meyn,et al.
Probabilistic Methods for Air Traffic Demand Forecasting
,
2002
.
[10]
G. W. Snedecor.
Statistical Methods
,
1964
.
[11]
Scott Smith,et al.
A New Model to Improve Aggregate Air Traffic Demand Predictions
,
2007
.
[12]
Craig Wanke,et al.
Incremental, Probabilistic Decision Making for En Route Traffic Management
,
2007
.