Automation in Air Traffic Control
Before the days of radar and computer based systems, Air Traffic Control was dependent entirely on Procedures. The importance of calculations to determine distance and time as well as techniques still communicated within pilots and ATC today such as position and altitude reporting.
Progress strips were passed one from one controller to anther (a practice which can still be found today)
However the 1950’s saw early attempts with introducing automated systems into air traffic control with the expansion of the air carrier system. One can say the 1970’s was generally the time period where computer systems were introduced into air traffic control, however this introduction came at different rates in different parts of the world.
Using the example of the United States, one of its first Computer Based ATC system was the IBM 360 which was modified as the IBM 9020, this was installed in all 20 EN Route centre throughout the country and was followed by Univarc’s ARTS III and ARTS II (minicomputer) for services as terminal facilities.
The functioning of this system is known as a closed loop control theory. Where by the system is self adjusting to the multitude of information that is it receiving, in other words receiving feedback and adjusting its method as compared to an open loop system where no feedback is given. Specifically for ATC, how the user wants the system to be configured at future points depends on the objectives at the planning stage. This can involve safety parameters that need to be input into the system such as separation criteria. Any excursions or deviations from this set of criteria are corrected by a controlling function through the feedback function. One can see that communication with other system elements such as aircraft, other ATC units and ground vehicles is key.
In addition to these variables, there are other factors that need to be considered, an example being the external forces of weather variables and the unreliability of both the human and non human components, which can be difficult to predict.
Early automated ATC systems were largely based on data collection and storage, which was then presented to the human being for interpretation. Later advancements in this automation technology have allowed certain data to be manipulated and displayed in different ways such as track deviations, and conflict alerts. One could term these systems as “automatic support” as humans still retain overall planning, problem solving and decision making control. This had an impact on the stress level that human air traffic controllers were experiencing.
Melton and colleagues performed psychological and biochemical assessments of ATC personnel in two American Facilities before and after the installation of the ARTS III system (Automated Radar Terminal Systems III). This system was capable of providing aircraft identity together with position, height, and movement information. This system was designed to reduce the amount of controller co-ordination required. Although human-machine interaction time increased (with keyboard entry tasks), overall workload was found to have decreased. But, even though it was hypothesised that the system would decrease stress (as measured through psychological measure and questionnaires), overall stress was found to have increased.
In 1980, a study of computer aiding was conducted at the University of Aston in the United Kingdom in conjunction with the Royal Signal and the Radar Establishment. The study examined an ATC simulation using computer assisted approach sequencing (CAAS) and London Heathrow, designed to assist in regulating inbound traffic at the airport, and interactive conflict resolution (ICR), which was designed to aid in the detection and resolution of en route traffic conflicts. The results were generally favourable and also resulted in a new method being used where inbound and outbound traffic was handled independently. This resulted in a better utilisation of the OCR system. World load was judged to be generally acceptable however concern was expressed that “computer aid might reduce demand on the controller to below an optimum level”.
The results of these two studies did not provide strong evidence that automation will be an increase or a decrease to controller stress. However, at the time of the study, the ARTS II system had only been in place for 5 months and the controllers were still not fully familiar with the system at the time of the experiment. The study was also conducted in a short space of time (a few days rather than months); where by a much longer period of time would be required when studying trends in non-routine circumstances. In addition, changes to stress levels can vary from subtle to chronic depending on the controller and his/her genitive ability, therefore any psychological or biochemical changes would be difficult to measure as a whole.
However with these limited studies, one cannot assume that it is the overall trend for automation in ATC for this specific time period. So far this article has presented studies that recorded incremental changes to a traffic controller’s work with automation as a support, not radical changes through advanced automation. Even with the UK experiment, some controllers reported unease at being so called “driven” by automated systems without being fully in charge of the situation. Some others also questioned whether having a mental picture would long survive a failure of computer equipment.
New automated systems may be able to simulate new practices and situations for better or worse; however human response and team performance still prove to be difficult to model beforehand.
Field stated in 1985 that “the computer is there to assist (the controller) in his task of safe and orderly flow of air traffic, and must in no way go beyond the capability of the human being to retain control of decision making”. In today’s practices, the air traffic controller still retains planning, analysis and control functions while the computer plays the supporting role. Even though this is only a supporting role, it still plays a very important part so much so that there are backup systems such as the US Direct Access Radar Control (DARC) are integral parts of the system. However one must note that the human must still maintain control of decision making. This is because of a variety of reasons:
1. Present systems still require the human to have a mental picture of air traffic and a mental model of airspace.
2. Controllers need to step in at any time should there be a computer failure
In many places, old manual methods still exist within modern systems and are still the final backup in case of any failures.
However as we advance in technology available, advanced automation systems do allow overall planning, analysis and control functions to be handles by the computer rather than the human. An example of such a system is the Long Range Radar and Display System III (LORADS III) in use at Singapore’s Changi Airport. Examples of Capabilities of future advanced systems include clearing multiple aircraft along individualised pilot request on “customised routes” according to weather, other aircraft and traffic flow.
One downside to this is that such a system would make it more difficult for the controller to maintain a mental situational model in his/her mind and would make it difficult for a rapid manual intervention if anything goes wrong.
In recent advancements, we see this challenge being taken on. One example being the FAA’s Advanced Automation System (AAS) with the idea of a single integrated en route and terminal service by controllers in tower cabs, enroute centres and other facilities. Instead of the system providing information for the controller to use, the controllers will be responsible for the general health of the automated system by providing it with information, except when dealing with exceptions and/or handling system failures. Unfortunately this system was terminated in 1994 with the admission that the project was probably too immense and unmanageable to begin with.
Boredom could again be a problem with this system and the professional atmosphere of a team is hard to be replaced by a system which is based largely on a one to one human to machine dialogue.
In conclusion, there is a clearly a future for advanced automated systems to increasingly take on a larger and more complex responsibilities in air traffic control. However while the general public, employers and controllers might appreciate the potential of automated systems in ATC, some controllers might perceive themselves to be servants to an impersonal system. This could lead to some dissonance. A human being who feels that he/she can exercise authority over his/her environment will feel less stress than one who has little or limited control. While automation takes away some of the control from the human being, on the other side, automation may reduce workload and increase attention across a wider scale. Whether automation reduces or increases stress depends on the situation and depends on a variety of details such as the managerial and technical structures within a certain ATC organisation. Whether automation will continue to increase its role in the future of ATC remains to be seen.
George, P., (2010) Open vs Closed Loop Systems, Retrieved 10th September from http://www.ehow.com/about_6073255_open-closed-loop-motor-control.html
Automated Radar Terminal System (ARTS) (2009) retrieved 10th September 2010 from http://www.faa.gov/air_traffic/technology/tamr/arts/
Edward, C., (2002) The Ugly History of Tool Development at the FAA Retrieved 10th September 2010 from http://www.baselinemag.com/c/a/Projects-Processes/The-Ugly-History-of-Tool-Development-at-the-FAA/
IBM 9020 Retrieved 11th September 2010 from http://www.thegalleryofoldiron.com/9020E.HTM
Data Communications (Data Comm) Retrieved 11th September 2010 from http://www.faa.gov/about/office_org/headquarters_offices/ato/service_units/techops/atc_comms_services/datacomm/
Rasmussen, J., Pejtersen, A.M., and Goodstein, L. (1991), Cognitive Engineering: Concepts and Applications, Vol.2: Applications. Working paper
Stokes, A., Kirsten. K., (2009), Flight Stress: Stress, Fatigue, and Performance in Aviation. Surrey, England: Ashgate Publishing
Hunt, V.R., Zellweger, A., (1987) Strategies for future air traffic control systems, Computer. Pages 19-32
Tattersall, A.J., Farmer, E.W., and Belyavin, A.J. (1990) Stress and Workload Management in Air Traffic Control in Stokes, A., Kirsten. K., (2009), Flight Stress: Stress, Fatigue, and Performance in Aviation. Surrey, England: Ashgate Publishing. Page 314
Field, A. (1985), International Air Traffic Control: Management of the Worlds Airspace. Oxford, England: Pergamon Press
Hopkin V.D (1991) The Impact of Automation on Air Traffic Control Systems in Stokes, A., Kirsten. K., (2009), Flight Stress: Stress, Fatigue, and Performance in Aviation. Surrey, England: Ashgate Publishing. Page 314
Melton, C.E., Meckenzie, J.M., Smith, R.C., Hoffman, S.M., and Saldivar, J.T (1976) Effects of ARTS-III, Aviaiton, Space and Enviromental Medicence, Vol. 47, Pages 925-930
Want to know more?