1. Training im DYSIM Lab
Die Trainingsmethoden für Air Traffic Controller werden in der regelmäßig um aktuelle Technologien erweiterten FAA Order JO 3120.4, Air Traffic Technical Training, beschrieben. Ein Vergleich der Order Stand 1986, JO 3120.4G, mit derselben Order Stand 2015, JO 3120.4P, zeigt diverse Fortentwicklungen und neue Techniken.
Zum Zeitpunkt 9/11 war die Order 3120.4J vom 16. Juni 1998 gültig. Ihr Nachfolgedokument ist JO 3120.4L vom 22. Juni 2005, das den folgenden Ausführungen zugrunde liegt.
Abschnitt fünf beschreibt den Umfang des Trainings für Controller eines ARTCC vor dem Radarschirm:
SECTION 5. STAGE IV: RADAR CONTROLLER TRAINING
PREREQUISITE: Successful completion of Stage III (Nonradar/Radar-Associate Controller Training).
CLASSROOM/SITUATIONAL TRAINING: This training is administered using FAA Academy-developed and facility-developed course materials for instruction of ATC procedures. This academic component of training consists of classroom instruction and adequate practice using CBI [Computer Based Instruction] and/or DYSIM [Dynamic Simulation] exercises.
SIMULATION TRAINING: This training consists of DYSIM laboratory time to administer the necessary familiarization, instructional, and evaluation scenarios.
OJT [On-the-Job-Training]: After successful completion of classroom training, OJT shall be conducted in the operational environment in accordance with Chapter 3 of this order.
Training mit Simulationen findet demnach mittels des Systems DYSIM [Dynamic Simulation] statt, das zwischen 1975 und 1977 entwickelt wurde. DYSIM ist “a high fidelity simulation facility that closely resembles the real en route ATC environment” (FAA-Report Reduced Posting and Marking Of Flight Progress Strips, S. 4). Die Funktionsweise des Systems erklärt die einschlägige Fachliteratur (Wickens, Flight to the Future, S. 73):
The FAA is currently using the operational computers in the air traffic control facilities (HOST and ARTS) to provide simulation training. In the en route centers’ simulation, it is called DYSIM, and in the terminal facilities, it is called ETG (enhanced target generator). Both systems generate simulated targets that can be maneuvered by a pseudo-pilot operating from a remote radar display with a keyboard. In these simulations, the radar display and keyboards are the same as those used in the actual control room, thereby creating what is called full-fidelity simulation.
Die in Order 3120.4L definierten Trainingssituationen umfassen hierbei alle denkbaren Aufgaben eines ATC, z.B.
(1) All radar identification methods and radar termination.
(2) Vectoring (e.g., to geographical point, to final approach course, for separation, departures, off route, around weather, no-gyro, flight breakup, sequencing).
(3) Departures and arrivals simultaneously in sector.
(4) Separation (e.g., overtaking situations; crossing, converging, and opposite direction traffic; from adjacent airspace, obstructions, and special use airspace; primary to primary, beacon to beacon, and beacon to primary; radar and nonradar).
Einige Übungsprofile geben einen Einblick in das DYSIM-Training: Beispiel 1 (1988, S. 3ff.), Beispiel 2 (2000, S. 5f.).
Das DYSIM Lab, in dem das Training stattfindet, ist dem regulären Arbeitsbetrieb eines ARTCC nachempfunden: Radarschirme, PCs, Controller, Vorgesetzte (Bildquelle, S. 6).
Im Unterschied zum OTJ-Training zeigen die Radarschirme allerdings keinen Realverkehr, sondern simulierte Radarsignale [inserts].
Radardaten sind die Daten bestimmter Radarstationen (Sekundär- und/oder Primärradar). Sowohl Radardaten auf dem Sekundär- als auch auf dem Primärradar setzen ein Objekt voraus, dass diese Daten durch Radarecho oder Transponder generiert.
Simulierte Radarsignale sind dagegen weder als Primär- noch Sekundärradarsignal sichtbar, da sie per definitionem nicht auf einem umherfliegenden Objekt beruhen – dieses Objekt wird ja nur auf dem Radarschirm simuliert. Sie werden daher auch nicht durch eine Radarstation generiert, sondern im jeweiligen Sim Lab der entsprechenden zivilen oder militärischen Einrichtung – im Fall eines ARTCC im DYSIM Lab.
In den Jahren seit der Entwicklung wurden Optik und Verhalten des DYSIM immer mehr an die Realität angeglichen. Ein Schnappschuss aus dem Jahr 1988 (Office of Technology Assessment, Safe Skies for Tomorrow, S. 146):
Developmentals training for ARTCCs undergo radar training with simulations created through the hardware and software of a system called Dynamic Simulation (DYSIM). DYSIM ties into the operational computer of the ARTCC, presenting the trainee with simulated traffic of limited realism. The Seattle ARTCC uses an upgrade of DYSIM, called Computer Enhanced Radar Training (CERT), which has improved software that more realistically simulates sector traffic. The CERT program emphasizes good use of instructors and the proficiency of operators who simulate pilots. As a result of these improvements, the Seattle ARTCC has cut by over 50 percent the time needed at certain stages of controller training, and has reduced time spent for on-the-job training with live traffic by 18 percent. The realism of training is limited, because software does not allow simulated traffic to deviate from preferential arrival and departure routes in the livesystem, and the call signs of simulated traffic must begin with XXX, a pattern never encountered with real traffic.
Ein Schnappschuss aus dem Jahr 1997 (Wickens, Flight to the Future, S. 73f.):
Scenario scripting is done by outlining the path of each aircraft in a specific simulation syntax that is difficult to master. This development process is extremely time-consuming. When the simulation is operating, it cannot be stopped and replayed for lesson reinforcement. In the ARTS ETG, in some instances the computer capacity of live traffic will limit the number of simulated targets available or even drop the simulated targets in the middle of a training session.
Ein Schnappschuss aus dem Jahr 2000 (FAA Office of Aviation Medicine, Reduced Posting and Marking Of Flight Progress Strips, S. 4):
Workstations are fully functional and landline communications are provided. Flight plans for each aircraft in a scenario are pre-programmed, but controllers or simulation pilots may change any flight plan during the simulation. Training specialists execute ATC instructions to simulate the roles of pilots and other controllers during the simulation.
Ein Schnappschuss aus dem Jahr 2007 (MITRE/FAA, Application of Advanced Technologies, S. 2-3f.):
En route field training consists of a combination of classroom, simulation, and OJT [On-the-job-Trainig]. OJT constitutes the major portion of field training.
FAA Order 3120.4 provides guidance on the conduct of OJT, but there is a great degree of variance in the time incurred. Several factors influence this length of time:
• Trainee aptitude and motivation.
• Evenness in the flow of incoming trainees. (Classes may be delayed until a minimum number of trainees are available.)
• Use of qualified trainees to staff sectors, thereby placing their training on hold.
• Degree of “seasoning” time required on certified sectors to build operational awareness and confidence before continuing to more complex and busier sectors.
• Effectiveness and quality of OJT time, which is highly dependent a traffic level and complexity that is appropriate for training.
DYSIM was deployed in the 1970’s and enhanced when possible over the intervening years. Given the current state of this technology, there are several opportunities for improvement:
• Increasing the realism of altitude and speed profiles for simulated aircraft
• Ability to account for wind effects on aircraft maneuvers
• Control of speed adjustments for simulated aircraft
• Reducing the dependence on human “pseudo-pilots” for scenario runs
• Ability to pause/resume a scenario run
• Ability to replay a scenario run for after-action review
As these types of improvements are made in future simulation capabilities, training will be able to rely more on simulation and less on OJT. Moreover, many of the research principles that promote improvements in training processes can start to be applied.
Bei allen Ähnlichkeiten wird jedoch eine grundsätzliche räumliche Trennung von Realität und Übung aufrechterhalten. DYSIM findet in einem eigenen Raum und auf eigenen Radar- bzw. Simulationsschirmen statt.
Das Training via DYSIM wird von den Controllern der 9/11 beteiligten ARTCC´s bestätigt (vgl. auch die Stellungnahmen zu “wargames”). Richard Beringer, ZBW:
Beringer considers hijackings pre-9/11 as extremely out of the ordinary, even though they are trained in the basic ATC response on a yearly basis. Beringer believes that multiple situations are practiced in the DynSim and CBI training, and also believes that a situation with the multiple factors that occurred on 9/11 has been practiced. […] Overall Beringer believes the FAA and military coexist well in their shared airspace. Beringer stated that he has handled fighter aircraft in FAA DynSim training.
Daniel Bueno, ZBW:
Bueno stated that the Dynamic Simulation Training (DynSim) that ATCs are required to perform yearly serve their purpose, even though they are only simulation. He noted that in the past one of his DynSims might have involved vectoring an aircraft toward a hijack, but if so it is only a loose memory, but that he definitely has not exercised a NORAC hijack with no transponder. Nor had their been a hijack simulation or exercise that included FAA and NORAD co-participation.
Shirley Kula, ZBW:
Kula noted the Dynamic Simulation (DynSim) training programs usually have a hijack scenario every year, but those scenarios in her experience have never consisted of multiple hijacks, or in her experience of a single hijack that necessitates a vectored military fighter.
Charles Alfaro, ZNY:
The training he received was computer refresher training and dynamic simulation exercises. He does not recall any multiple hijacking exercises. Further, Alfaro noted that all his training dealt with single-event scenarios.
Lorraine Barrett, ZNY:
Pre 9-11 there was hijacking content in exercises in the Dynamic Simulation lab, but they were not extensive exercises. Training was vastly different than the actual hijacking of airplanes on 9-11. In training scenarios the pilot was still in control of the airplane. All exercises were based on that given. There were also scenarios when the pilot would use a code word or the pilot would verbally explain if the situation allowed.
David Bottiglia, ZNY:
Training on Dynamic Simulations involves airplanes that squawk hijack code 7500, inferring that the airliner pilot is in control of the aircraft. Bottiglia continued his point by noting that controllers were never presented with a scenario or practice exercise that was more challenging than this. The procedure that he had learned was 1) to see the “hijack” warning flash on the scope; 2) to verify with the pilot that he ist’squawking 7500″; and 3) then to tell the supervisor for the area. Bottiglia noted to Commission staff that he has never been involved in a real life military intercept on a hij ack and has never participated in a simulation that would vector a military aircraft towards a target.
George Leonard, ZNY:
He had never contemplated a scenario such as 9-11. All employees undergo annual refresher training internal to FAA that included a hijack scenario at least once per year. The Dynamic Simulation lab where scenarios are created was intended for controllers. He had no recall of unusual scenarios; most training scenarios were standard hijackings.
Mark Merced, ZNY:
Merced’s hijacking training was mostly annual refresher training that included hijack scenarios on a Dynamic Simulation, called DynSim, which simulates air traffic. These exercises, as noted, involved one aircraft, and a pilot as the participant in communication through code use. Even post 9/11 Merced has only exercised scenarios involving single aircraft.
Anthony Palmieri, ZNY:
Palmieri is trained once a year in a Dynamic Simulation (SynSim) machine with hijack simulations.
Paul Thumser, ZNY:
He was not familiar with Dynamic Simulation training concerning hijackings and had no computer or other training for hijacking. Operations supervisors do not go through the same training as controllers.
Chris Tucker, ZNY:
Tucker commented that he received refresher training quarterly. They were told not to do certain things. The training was on the DynSim simulator. He commented that there training did not entail 9/11 scenarios; but that the training does encourage a controller to react the best they can under different problems.
Linda Povinelli, ZID:
Povinelli had never before been involved in a hijack situation. She has participated in refresher training in the DYSIM, dynamic simulator about every six months. Povinelli had never been presented with a suicide hijack situation in this training.
Linda Justice, ZOB:
In simulation labs representing hijack scenarios, the controller has to look for them to say “trip” and squak a different mode. “Squak” or “trip” are signals to the controller of a hijacking. Deviating from the course or transponder switch was never incorporated into the simulation.
Brudnicki, Dan (MITRE)/Ethier, Bob (FAA)/Chastain, Kerri (FAA): Application of Advanced Technologies for Training the Next Generation of Air Traffic Controllers. MITRE 2007 (Online)
Office of Aviation Medicine (FAA): Reduced Posting and Marking Of Flight Progress Strips for En Route Air Traffic Control. Washington D.C.: Office of Aviation Medicine 2000 (Online)
Office of Technology Assessment: Safe Skies for Tomorrow. Aviation Safety in a Competitive Environment. Band 22. Kap. 7. United States Congress 1988 (Online)
Wickens, Christopher D. et al (Ed.): Flight to the Future: Human Factors in Air Traffic Control. Washington D.C.: National Academies Press 1997