Development of Airspace Sector and Encounter Models to Support the Analysis of Aircraft Separation and Collision Risk (1998)

This report presents a first-order analysis of blind conflicts expected to affect the NAS system in the near future under two Free Flight operational concepts: RVSM and Cruise Climb. The study focused on the development and use of two computer models (AOM and AEM) to respectively predict traffic flows across well defined volumes of airspace, and the number of potential blind conflicts if all flight plans are executed without controller or pilot intervention. The models developed have been coded in Matlab, a general engineering language, facilitating their execution on any computer platform (PCs, PowerPC Macs, and UNIX workstations) without modifications.

While this study provides a first-order approximation of the level of conflict exposure in a particular center or sector it does not provide a measure of collision risk in the true sense. Further investigation of the end-game ATC controller and pilot dynamics (including aircraft navigational accuracy) is needed to truly quantify collision risk.

Some insightful computational test are conducted to understand traffic pattern variations and blind conflicts in four enroute control centers in CONUS. The time and spatial characteristics of these conflicts were studied using the tools developed to provide a view into the type of conflict encounters expected in future NAS operations. The hope is that these tools would be further refined to assess collision risk incorporating human and vehicle reliability models.

Several conclusions can be derived from this case study:

1) There would be likely moderate to substantial variations in traffic flow patterns across various ARTCC sectors in NAS. The introduction of flexible flight planning rules expected in Free Flight would affect differently various ARTCC centers according to their geographical location. In this study ZMA and ZJX centers had less variation in 15-minute traffic flows than those observed across ZID and ZTL.

2) The number of potential conflicts in the enroute airspace system would decrease with the introduction of Free Flight operations if reduced vertical separation criteria is allowed. It is not possible to quantify the risk associated with reduced separation blind conflicts using the models developed. However, further investigation is needed since ATC controllers and pilots operating under RVSM rules might have less time to react to blunders under these circumstances (assuming current levels of automation).

3) The number of blind conflicts expected under Cruise Climb and RVSM modes (as defined in this report in Chapter 5) are of the same order of magnitude. It is not clear how ATC controllers would react to potential conflicts between two or more aircraft operating in a cruise climb and what would be their influence on collision risk. Further investigation is necessary.

4) In general, there are substantial to moderate differences in the time and space distribution of blind conflicts under RVSM and Cruise Climb scenarios. The effect of these distributions in ATC controller monitoring workload and eventual reliability to intervene under blunder conditions should be further investigated.

5) In general, vertical transition conflict times under RVSM and Cruise Climb scenarios are expected to be shorter in duration due to the smaller vertical separation criteria. Enroute conflict times (i.e., coplanar conflicts) varied significantly. Under some circumstances, enroute conflict times increased for at least one of the Free Flight scenarios investigated.

6) The distribution of relative headings of conflicts varied in the transition to some Free Flight scenarios (i.e., cruise climb). This parameter could have important implications on how controllers perceive conflicts and eventually, on the intervention modes used to separate traffic. Further investigation of this important parameter is also needed.
[Full Text]