We have all heard of the so called airport hot spots… places where extra caution is required to avoid nasty incidents like runway incursions. Until now, hot spots were discovered the hard way. Usually the dangerous places were identified as such following several incidents that made the situation clear: this is where danger lurks, extra caution advised. For the same reason, it is difficult to engineer out hot spots even in green field projects as it is not easy to establish before the start of operations just where things will consistently go wrong.
The Airport Viewer being developed by the FAA and Saab Sensis Corporation will change all that. A system originally developed to collect operational data to be used to judge the effects of various NextGen elements on airport operations is turning out to be a powerful tool to assess otherwise hard to notice operational anomalies which can lead to serious ground movement incidents.
Like in so many other areas, the key to this potential safety improvement is the harvesting and processing of ground movement data that has always been there albeit in a form that did not lend itself to easy interpretation. ASDE-X, the Airport Surface Detection Equipment, Model X (which is also a Sensis product) is being deployed across the US at all major airports. It monitors and records live traffic which the FAA can review like data from any other of their surveillance systems.
In the past, the recorded data was reviewed usually only when an incident was being investigated. After the fact as it were. Yet the circumstances that led to many incidents were there also in the past, possibly hiding in the mass of data. Even if things came very near to being an incident, if it did not happen, the almost-event went unnoticed.
Sensis is now creating a few clever algorithms which, let loose on the ASDE-X data, are able to discern movement patterns and behaviors which represent anomalies and which may indicate problem areas. Possible hot spots!
The idea to mine ASDE-X data was a natural, after all data mining is helping organizations also in other industries to improve their processes and activities. The idea to look for ground movement related anomalies with a view to identifying potential risk areas was the new element.
Using data samples from several major airports, engineers first defined what could be considered as the normal behavior at a given airport. Like tracks in the air, ground movement tracks also tend to average out into several typical patterns and a large enough sample will serve as a “signature” for that airport. In part using live traffic samples but also using expert judgment, a catalogue of anomalies was established. A deviation from the usual route, sudden deceleration, abnormally long waiting, unusual turns, etc. can be taken as signs that something is deviating from the normal course of events. One or two such anomalies may not indicate a problem but a multitude of them is a sure sign of confusion, incorrect procedures, poor taxiway design and so on. The algorithms also look for trends. For instance if at an area that had not problems before things suddenly start to go wrong, the question needs to be asked: what has changed that causes the new anomaly?
Clearly by not waiting for an incident to happen but acting when an anomaly is discovered can prevent more serious problems.
The types of anomalies to consider are relatively limited but their correct recognition will improve over time as more and more of them are picked up and analyzed. The airport “signatures” will also be refined over time and it will be interesting to see how signature elements across multiple airports will have eventually similar anomalies associated with them. This will be a huge help to airport designers in avoiding hot spots from the ground up.
Enhancements to the Airport Viewer will include synched controller/pilot voice transmissions (and I guess data link messages in due time) and also extending the data capture to the terminal area enabling an even more complete picture of the behavior of aircraft approaching the airport in question.
The Airport Viewer is an excellent example of the power of data if it is collected and properly analyzed. The safety benefit that is the aim of the current development work is possibly the most important but surely not the last type of use ground movement data can be put to.
Other applications will be developed in due time to benefit not only safety but also efficiency.