NASA's research into a takeoff performance monitoring system (TOPMS) originated as the doctoral thesis of a Kansas University student named Raghavachari Srivatsan. Srivatsan believed the number of takeoffrelated accidents could be reduced if pilots were given more information on the airplane's progress throughout the takeoff roll. In 1984, Srivatsan began working with researchers at the Langley Research Center on display concepts that might be able to give pilots this kind of information.

The basic TOPMS information was designed to be displayed on the navigation CRT screen in the cockpit. The pilot would input the airplane's weight, the outside temperature, wind conditions, runway length and the basic runway condition (dry, wet, icy, snowcovered, etc.) into the flight management computer. The computer would then create a display of the departure runway that showed the predicted takeoff point of the airplane, the point at which the airplane had to rotate in order to clear a 35foot obstacle at the end of the runway, and where on the runway the critical reference speeds should occur. The display also incorporated two moving bars that indicated the engine pressure ratio (EPR) for each of the two engines. If either or both of the engines failed during the takeoff roll, the corresponding EPR bar would turn red, alerting the pilot to the engine problem. The airspeed was displayed in digital form in a box beside the runway.

Three seconds after the pilot stopped advancing the throttles on the takeoff roll, the computer would compare the predicted acceleration of the airplane against the actual movement its onboard accelerometers were reporting. If the two rates did not agree, the computer would adjust its equations and show a new predicted takeoff point on the TOPMS display. During the takeoff roll, the computer would also be running a series of tests on the airplane's condition and modifying the display accordingly. The first test was whether or not both engines had failed. If they had, not only would both EPR bars turn red, but a red stop sign would appear at the top of the display. If both engines had not failed, the computer went on to the next test, which was whether or not the airplane had reached the critical decision speed, known as V1. Once the airplane had passed V1 speed, it was committed to the takeoff, and a large green arrow would appear at the top of the screen, indicating to the pilot that the takeoff should be continued.

If V1 speed had not yet been reached, the computer tested to see if one engine had failed. If an engine failure was detected before the airplane reached V1 speed, a red stop sign would appear on the screen, advising the pilot to abort the takeoff. If both engines were still operating, the computer next evaluated the acceleration of the airplane. If the acceleration rate was below a predetermined threshold, a red stop sign appeared on the display. If the acceleration rate was still within acceptable limits, the computer looked to see if the predicted takeoff point had progressed beyond the point on the runway where the airplane could rotate and still clear a 35foot obstacle, even if one engine failed. If it had, the stop sign appeared. The final test was to see if the predicted stopping point of the airplane would still be on the runway, if the pilot aborted. Once the stopping point was beyond the runway, the display showed a green arrow at the top of the screen. If the pilot aborted the takeoff at any point, all the takeoff information disappeared, and the display indicated only the predicted stopping point of the airplane at its current deceleration rate and with maximum braking efforts. [Ref 6-15]

Most of the TOPMS research was actually performed in the TSRV fixedbase simulator at Langley. The first simulator study, conducted in 19851986, consisted of 32 pilots from NASA, the FAA, the United States Air Force, airline companies, and airframe manufacturers. The pilots liked the display, but were concerned about the pilot looking down to monitor the information, instead of looking out the window. As a result of the study recommendations, the NASA researchers organized a second simulator study in 19861987, in which 17 pilots were asked to evaluate the TOPMS on both a "headsdown display" (HDD) and a "headsup display" (HUD). The pilots rated the headsdown display as "good" and the headsup display as "very good." All the pilots reported, however, that they would like to have at least a headsdown TOPMS display in their cockpits, because the information provided "valuable safety information not currently available during takeoff." [Ref 6-16]

Following the second simulator study, TOPMS was installed on the TSRV 737 to evaluate how well it would operate under actual flight conditions. In April, 1988, the ATOPS crew took the airplane down to the Kennedy Space Center to test the system on the same 15,000 foot runway the Space Shuttle used for Florida landings. The researchers wanted to test the accuracy of the system's predicted stopping points for highspeed takeoff aborts, and the long, wide runway at Cape Canaveral provided a reassuring extra safety margin. After the predictions proved accurate during 80 knot and 100 knot aborted takeoff runs in Florida, the crew conducted similar experiments back at the Wallops Island flight facility in Virginia. Between other research projects, maintenance and scheduled upgrades to the 737's experimental systems, however, only six days of TOPMS flight testing were completed over a two year period. In addition, all the test runs were on dry runway conditions. [Ref 6-17]

In October 1989, NASA held a workshop on TOPMS that was attended by representatives from manufacturers, airlines, avionics companies, flight safety groups, and industry associations. Despite the positive reviews by the test pilots in the simulator studies, the industry response to the concept was mixed. One concern was that the sudden appearance of the "STOP" and "GO" flags on the display might be interpreted by pilots as a command instead of an advisory caution. Another concern was that the predicted information on the display might not always be accurate, because there were too many potential variables. For example, conditions could change from one end of the runway to the other, so although the display might indicate to the pilot that he had enough room to stop, that might not actually be the case. [Ref 6-18]

To try to address some of the concerns raised at the workshop, Srivatsan and the Langley researchers conducted a third simulation study with six pilots in 19911992. This study evaluated three different TOPMS display options. The first showed a predicted takeoff point but had no advisory symbols and did not include information on the predicted stopping point in the case of an aborted takeoff. The second option was the original TOPMS display, and the third contained the same predictive and advisory information as the second option, with one additional feature. In the third option, the appearance of an preliminary abort symbol indicated a developing acceleration problem before a firm "STOP" advisory appeared on the screen. Four of the pilots preferred the third display option, while two preferred the format without any advisory information, although all the pilots reported that they would prefer any of the options to no TOPMS display at all. [Ref 6-19]

The different display formats still did not address the industry concern of runway variables, however, and one other significant problem remained. In manufacturing a warning system to be used in circumstances where the consequences of a missed alert could be serious, such as in a critical takeoff situation, the alert tolerances had to be made very tight. Tight tolerances could easily lead to nuisance alerts, however. Nuisance alerts would not only prompt unnecessary takeoff aborts, which were in themselves somewhat hazardous, but they could also lead pilots to pay less attention to alerts when they were triggered. As a result, the pilots might ignore a warning in a situation that was actually critical, negating the value of an alert system that had cost a lot of money to install. [Ref 6-20]

In short, safety from a manufacturer's or operator's standpoint was not as simple as it might appear to research organizations. [Ref 6-21] TOPMS seemed to give pilots valuable safety information, and the test pilots who participated in the demonstration all thought it was a beneficial system. But TOPMS would need a lot more work before it would be acceptable to the airline industry, which had to take different and more complex factors into account. As a result, Boeing elected not to include a takeoff performance monitor in the B777 design. [Ref 6-22] McDonnell Douglas, on the other hand, saw at least enough potential in the technology to continue internal research on the concept. Based on the expectation that additional research might be able to overcome the technology's perceived deficiencies, McDonnell Douglas also included a takeoff performance monitor in the tentative baseline design for the MDXX cockpit. [Ref 6-23]