Three studies were conducted to determine pilots' performance in responding to advisories given by the Traffic Alert and Collision Avoidance System (TCAS II). Pilot responses were evaluated to normal TCAS II resolution advisories (Study One), to amended advisories (Study Two), and to alternative display formats (Study Three).
In the first study, normal TCAS II operations were evaluated in simulated air carrier line operations. Sixteen three-person airline flight crews, currently flying the Boeing 727, served as subjects. Each crew flew eight flights with or without TCAS as part of the full-mission simulation. Pilots' performance of the avoidance maneuvers and their evaluation of the system were measured.
When the TCAS was in use, no other aircraft came within 200 ft vertically and 1000 ft horizontally in 96 flight segments. In 32 flight segments without TCAS, there were four instances in which minimum aircraft separation was less than these limits. All maneuvers in response to the 59 TCAS resolution advisories were made in the correct direction and initiated within the five seconds allowed by the system, except in one instance, which also resulted in safe separation. These results indicate that (1) pilots were able to utilize TCAS II correctly within the response times allocated by the system, and that (2) TCAS II was effective in ameliorating the severity of the simulated traffic conflicts.
The second study tested pilots' responses to proposed changes in the avoidance advisories. If a conflicting aircraft's path is projected too close to a TCAS-equipped aircraft, the system will issue an avoidance resolution advisory. If after a resolution advisory has been issued, the conflicting aircraft maneuvers in such a way that safe separation is no longer projected, the modified TCAS would issue a second resolution advisory. This advisory may be an increase in climb or descent rate from 1500 to 2500 ft/min, or a reversal from a climb to a descent or vice versa.
The proposed TCAS II logic changes assumed that the pilot would respond to the second resolution advisory in two seconds and use ±0.5 g acceleration. The second study measured pilot performance of these maneuvers. Ten Boeing 727 qualified airline pilots performed all the sequences of one to three resolution maneuvers under varying conditions of gross weight and altitude. The following questions were addressed: (1) Can the pilot detect the change in the maneuver advisory? (2) Can the pilot respond promptly and accurately to the new advisory? and (3) Can the maneuver be performed in the normal operating envelope of the aircraft?
The recorded reaction times suggest that pilots are able to respond within the two seconds targeted by the TCAS logic. The responses resulted in an acceleration which was less than originally assumed by the logic for the second and third maneuver (0.5 g assumed, 0.41 and 0.38 g measured). The increase in rate from 1500 to 2700 ft/min was successfully accomplished in 85% of the cases. The reversal maneuver, changing from climb to descent or descent to climb, was only successful in 15% of the cases. With 10 sec or more, the pilots were successful in reaching the prescribed vertical speed for 87% of the advisories. These data have been used to modify the TCAS II logic to reflect actual pilot performance. This will result in the issuance of safe and appropriate secondary resolution advisories in the rare instance when the conflicting aircraft maneuvers, and by doing so invalidates the initial avoidance advisory.
The objective of the third study was to examine pilot reactions to alternative versions of the resolution advisory display of TCAS II. Three display types were tested. The original TCAS resolution advisory display indicates with red lights which vertical speeds must be avoided. The pilot is told to get the vertical speed needle out of the red area. This study tested the hypothesis that the addition of a green “go to” target area, displayed on the vertical speed indicator, might be more effective than the “red-only” version currently in use. In addition, a third display containing only the green target area was compared.
Fourteen resolution advisories of each color pattern were randomly presented twice on a computer workstation to 36 volunteer pilots. The pilots responded to the resolution advisory by pushing (descending) or pulling (climbing) a computer mouse, or by doing nothing when the advisory did not require a response (a preventive resolution advisory). The results indicated that the red and green display resulted in less errors and quicker responses than did the red-only display. The pilots preferred the red and green display over the red-only display. To optimize pilot performance, the TCAS II display standard has been changed to require the green target area when pilot response is required.