PHARE Demonstration 1 Experimental Design

PHARE Logo
For PHARE Demonstration 1 three system ‘organisations’ (ORGs) were defined: a baseline (ORG 0); ORG 1, which examined the effect of the introduction of the PHARE Advanced Tools to assist the controller in implementing the PHARE Demonstration 1 operational concept of ‘advanced planning’; and ORG 2, which examined the effect of introducing aircraft equipped with 4-D FMS and datalink.

The objectives of the trial, namely to measure the impact of the PATs, 4-D FMS and datalink, were met by comparing the results from the various organisations whilst keeping other factors, e.g. controller role and traffic sample, the same.

The characteristics of the different organisations were as follows:

ORG 0

  • Mixed population, all 3-D Flight Management System aircraft, 3 traffic volumes, No PHARE Advanced Tools, Controller plans ahead based upon flight data, procedures to suit paperless system

ORG 1

  • Same traffic samples as ORG 0, Included the following PHARE Advanced Tools: Trajectory predictor, Flight path monitor, Conflict probe, Highly interactive problem solver (HIPS), ‘Advanced planning’; computer assistance looks up to 20 minutes ahead to design conflict-free trajectories, procedures to suit PHARE Advanced Tools

ORG 2 (30%)

  • Mixed population 30% 4-D aircraft, 1 live aircraft, 2 traffic volumes, same PHARE Advanced Tools as ORG 1, procedures as ORG 1 enhanced for 4-D Flight Management System and datalink aircraft

ORG 2 (70%)

  • Mixed population 70% 4-D aircraft, 1 live aircraft, 3 traffic volumes, same PHARE Advanced Tools as ORG 1, procedures as ORG 2 (30%)
The traffic samples used in the experiments were based on those previously used and validated for the NATS’ Computer Assistance for En-Route (CAER) trials. The samples consisted of projected ‘busy day’ traffic flows for the year 2000, and were generated by ‘growing’ a selected base day, Friday 17 July 1992, to the traffic demand forecast for the year 2000. These samples were then grown further by increasing the numbers of aircraft per route by 20% and 40% to provide ‘medium’ and ‘high’ volume traffic samples. Finally, the traffic demand was smoothed, so that predicted hourly capacity limits of the major airports were not exceeded, and a 75 minute ‘slice’ was taken from the day’s traffic. For each level of traffic volume, several samples were used to ensure the generality of the results and to prevent learning effects perturbing the results of the later simulation runs.

Each controller participated in one week of measured runs and performed the same role on the same sector in each run from that week. Fixing the role of the controller in this way meant that statistical tests could be used which allowed the response to be measured of each controller individually to the different organisations and it also reduced the effect on the comparisons of variability between controllers.

The number of runs that could be conducted during the PHARE Demonstration 1 trials was limited by time and controller availability. Mondays were used for refresher training, and during the other four days a total of fourteen simulation runs were carried out, scheduled according to the weekly timetable.

For further information

The design of the PHARE Demonstration 1 experiment is described in more detail in:
  Acrobat 96-70-24 PD1 Final Report Volume 2 Annex A Experimental Design and Methods58 Kb