controller to choose the preferred level of automation (management by delegation or by consent; Billings, 1997) and therefore the preferred personal level of involvement in the air traffic control task creates a cooperative human-machine system where the controller remains in the center of the system and is able to maintain a comprehensive level of awareness of the air traffic. The system may also be adaptable in the sense that the controller may communicate to the automation his/her information presentation, action, and decision-aid preferences when initiating control of the sector at the beginning of a shift just as in the Lockheed Pilot's Associate ( Banks & Lizza, 1991).
A recent study by Hilburn, et al. ( 1997) demonstrated the potential for successful implementation of such an adaptive information automation system. They tested air traffic controllers in a simulation of Amsterdam airspace using the Center TRACON Automation System (CTAS), an automated tool aimed at assisting air traffic controllers with managing and controlling arriving traffic into terminal airspace. CTAS is composed of three tools: the Traffic Management Advisor, the Descent Advisor (DA), and the Final Approach Spacing Tool. Hilburn et.al. used only the DA in this study to provide the controllers with estimated time of arrival for aircraft and a traffic plan to sequence and schedule the traffic into the terminal airspace. The automation detected arrival time conflicts and projected separation conflicts, and offered advisories aimed at solving these conflicts. Their results showed that when the automated system was operated on an adaptive schedule such that it was activated only under a high task load period, the workload of the controller as measured subjectively and physiologically decreased.
To further substantiate the need for an adaptive methodology, a study by Galster, et.al. ( 1998) demonstrated that if the ATM system evolves to a FF architecture and does not include adaptive information management as well as adaptive aiding tools, the controllers' workload levels will be impacted. The results of this study of simulated high traffic load revealed controller operational errors in detecting probable conflicts between pairs of aircraft as well as an increase in perceived workload.
The future of ATM includes the addition of automation for that of the air traffic controller. Assistance in the form of information automation as well as automated decision aid tools will be necessary for alleviating the extreme levels of workload predicted to occur for a controller working in a saturated traffic environment. In implementing this automation, designers must first consider the information needs of the controller and a system whereby relevant information is given to the controller at the appropriate time. An information manager that is both adaptive and adaptable is a possible solution.
It is also important to consider that novice and expert controllers are individual users of the ATM system and therefore it is necessary to design an interface for both types of users ( Hancock & Chignell, 1988). In considering both types of air traffic controllers, the interface will be able to ease the transition and training of controllers from the current system of air traffic control to the future system of air traffic management.
This research is supported by National Aeronautics and Space Administration Grant No. NAG-2-1096 from NASA Ames Research Center ( Kevin Corker is the program manager), and Grant No. NGT-1-52137 from NASA Langley Research Center ( Paul Schutte is the technical monitor). The views presented here are those of the authors and are not necessarily representative of NASA.
Banks S. & Lizza C. ( 1991). Pilot's associate: A cooperative knowledge-based system application. IEEEExpert, 6