INTRODUCTION. While the reliability of the modern jet engine has reduced the probability of an engine failure during takeoff the RTO continues to represent a significant safety hazard to civil aviation.

There are many situations likely to provoke an RTO, but it should be clearly understood that the V1 concept refers to engine failure only.

Unfortunately this simple fact has not always been clearly conveyed to airline pilots and many pilots have come to regard V1 as the go/no_go decision speed for any recognized anomaly during the takeoff roll, regardless of other factors.

HISTORY. Based on available statistics the odds for an RTO are:
· 1 RTO per 2.000 takeoffs.
· 1 RTO per crew every 4 years.
· 1 percent of RTOs occur near V1.
· 1 crew in 14 will experience an RTO near V1 in a 25 gear career.

The same source gives the odds for RTO incidents/accidents as:

· 3 in 25 RTOs result in an incident (runway departure)
· 1 in 25 RTOs result in an accident

There are many reasons offered for the bad history of RTOs: Unrealistic takeoff certification, failure to account for lineup distance, imprecise recommendations given by the manufacturer (to avoid liability), failure to understand the nature of tire failures, inadequate training of pilots etc.
We are not going to discuss the three former allegations, as we can do little or nothing about them. We will concentrate on the two latter, which we consider most important and which we are able to do something about.

FACTORS INFLUENCING THE RTO DECISION.
Type of failure, as perceived on flight deck
· Engine (flameout, fire, severe damage, FOD, vibration)
· Tire(s) (vibrations, gear indication malfunctions, hydraulic malfunctions)
· Vibrations from unidentified source
· Systems (master warning, master caution)
¨ Flight instruments
¨ Flight Controls
· Incapacitation

Of the factors listed above. two are predominant. A 10 year study of Douglas_built aircraft shows that for all RTOs
· Approximately 75 percent were due to tires and engine problems
· Of the 75 percent, tires accounted for slightly more than half

Other sources indicate that RTOs in response to tire problems are four times more likely to result in an accident than those in response to engine problems.

A number of factors related to performance will invariably influence the GO/no_go decision:
· Gross weight
· Runway conditions
· Runway margins (available RWY, area beyond)
· Obstacles
· Atmospheric conditions (thunderstorms, wind-shear, heavy precipitation)

CONSIDERATIONS. The decreasing occurrence of rejected takeoffs is an indication of improved overall systems reliability. This gradual improve-ment introduces a new element, we are seldom prepared for a maximum effort stop when a takeoff must be rejected from speeds near V1.

In general, we only encounter rejected takeoff situations in the flight simulator training during initial conversion to a new aircraft type and during recurrent training. Even then it is under relatively idealized conditions since our night simulators do not have the full capability to realistically represent those difficult go/no-go situations which arise from unusual vibrations, sound etc. In line operations we are oriented from the time we board the aircraft to "going" to the destination. In fact, the passen-gers onboard paid to do just that very thing. Probably this orientation causes an initial hesitation in making go/no-go decision. Any such hesitation at speeds approaching V1 will jeopardize the success of the stopping maneuver.

Typically at V1 the airplane rate of acceleration is about 3 to 5 knots per second with all engines operating. For every second that passes before a decision to stop or go is made, the speed of the airplane is increasing and approximately 70 meters of runway are used If the problem which is necessitating a go/no-go decision occurs on the low side but in the vicinity of V1, the combination of high acceleration rate, our state of mind and the probability of a more complicated set of circum-stances surrounding the decision than experienced in the flight simulator all tend to indicate that the airplane speed will be above V1 by the time the failure is recognized and real stopping procedures have been implemented.

By being predisposed to stopping, adequate thought may not be given to the meaning of V1 or airplane performance characteristics. V1 is defined by FAA rules as the speed at which an engine failure has been recognized and action initiated to either CONTINUE or STOP the takeoff. It is simply the speed at which a pilot changes his pre-planned response. The time to begin the decision making process is not at, or near, V1.

If we realistically look at the airplane acceleration rate around V1, the stale of mind of the crew, the fact that maximum effort braking is hardly ever practiced in normal operations and the fact that clearing slightly less than 35 feet at the end of the runway is not nearly as dangerous as running off the end of the runway, one might come to a con-clusion that on a runway limited takeoff the go decision may be better than the stop decision.

Engineering calculations would show that for many airplanes the height at the end of runway is reduced between one and three feet for each knot below V1 after an engine failure when the takeoff is continued. This shows the relatively minimal consequences of continuing the takeoff from a speed slightly below V1 as opposed to rejecting the takeoff from a speed above V1. We should not take such an extrapolation to an extreme, but there is , for most airplanes, a reasonable range around V1, where this is true. A little better understanding of the performance capability of the airplane in this most important area could be very valuable and would provide additional background to focus at-tention on the criticality of the go/no-go decision. On wet and slippery runways without rolling resistance our acceptance and use of reduced V1 is also an acceptance of a lower margin, down to 15 foot clearance over the runways end. (Engine fail-ure at reduced V1.) In order to further reduce recurring overrun incidents and accidents, perhaps the time has arrived to recognized that a small trade in clearance at the runway end is also ac-ceptable under other conditions on the remote occasions when a critical go/no-go decision has to be made.

TIRE FAILURES. As stated above we have in the past concentrated on takeoff problems related to engine failures. Given the statistics and the studies made, we feel it is time to devote more atten-tion to other factors involved.
Predominant of all other factors is, as stated earlier, tire failure.

What can happen to a tire?
· Loss of inflation
· Loss of thread
¨ With or without FOD
¨ With or without vibrations
¨ With or without damage to adjacent wheels

Why: Poor operating techniques:
· Fast taxiing
· Riding the brakes
· Excessive braking (The First Turnoff Syn-drome)
· Long taxi distances

Maintenance:
· Relaxed maintenance
· FOD to tires

PERFORMANCE WITH BLOWN TIRES.

NO GO. Obviously, with a damaged tire the stop-ping capability of the aircraft will be reduced, as there are E.G.:
· Loss of braking capability on damaged wheel
· Risk of further damage
¨ To other tires
¨ To hydraulic system

Hence, with a tire failure it is quite impossible to stop from V1 at a runway limited gross weight. If rejecting the takeoff, the situation may soon become aggravated and the decision cannot be re-versed.

GO. The consequences of a "go" decision seem less dramatic. There will be:
· Slight increase in rolling distance
· Slight reduction in height over runway end
 (These two facts are of no significance at all as long as engines are developing normal thrust)
· An unpredictable risk of FOD to an mounted engine
(you may not be able to verify a tire failure, but you will certainly be able to verify whether engines operate at normal thrust or not, and the status of the engines is the crucial infor-mation)

OTHER FAILURES. Statistics indicate there have been a number of takeoffs rejected at high speed for reasons really not justifying such action. A number of these rejected takeoffs resulted is inci-dents or accidents.

The exclusion of distracting elements during the critical time interval, i.e. from some point before V1 until some point after V1, has been given consid-erable attention in the past, but resulted in few practical solutions.

In an attempt by Boeing to reduce the possibility of a high speed rejected takeoff for unwarranted reasons, on e.g. 767 airplanes, warnings for system failure not critical to the takeoff phase are inhib-ited. The beeper and master caution lights are inhibited for all cautions after 80 knots and until either 20 seconds after liftoff or reaching 400 feet.

MALFUNCTIONING/ABNORMAL AIRCRAFT BEHAVIOUR DURING TAKEOFF. In order to aid in reaching a quick and correct decision should an abnormality occur, it is recommended that the PiC prior to starting a takeoff, makes a mental review of factors affecting that particular takeoff, e.g.:
· Gross weight
· Available runway length (overrun) and runway conditions
· Action in case of tire failure
· Obstacles (beyond the threshold and in the climbout area)
· Climbout conditions (icing, wind conditions)
· Aircraft serviceability and technical remarks

If the abnormality is seriously affecting the takeoff thrust, e.g. engine failure, the following basic rules apply:
· If occurring at or after V1, normally continue
· If occurring before V1, normally discontinue

Generally an aircraft can be expected to lift off at least at the end of the required runway length in case of a continued takeoff after engine failure at speeds as follows.
· 2 engine aircraft: V1 -10.
· All aircraft on wet and slippery runways when reduced V1 is used: V1 -5.

CAUTION: The basic V1 concept is based on dry runway conditions and since there is no full accountability for contamination, it will normally not be possible to stop the aircraft on a runway length- limited takeoff from speeds close to V1 when runway is covered with water, slush, snow or ice, even if reversing, V1 for wet and slippery runways and prescribed corrections have been applied.

When other abnormalities occur, their nature and time of occurrence must be taken into consideration.
· If occurring at an early stage of the takeoff where no doubt exists as to a safe stopping on the runway, then discontinue
· If occurring at speeds close to reaching V1, the nature of the abnormality and its effect on the airworthiness of the aircraft in a continued flight must be judged versus the possibility of making a safe stop.

The following type of abnormalities may justify a continued takeoff:
· Engine fire warning when all engines develop normal thrust
· Tire failure close to V1 on a marginal RWY with all engines developing desired power
· Indication failure of instruments not absolutely required
· General electrical failures
· Pilot incapacitation (body not blocking controls)
A takeoff discontinued at speeds above V1 on a minimum length runway is unprotected from a performance point of view.

For detailed flight deck procedures, see respective AOM.

Experience has proved that blown tires or landing gear structural failures may cause damage to landing gear doors, brake system, fuselage, wings and flaps as well as wiring and tubing in the landing gear wheel well. In cases of noticed or reported failures of this type, it is therefore recommended:
· To keep landing gear extended for at least 5 minutes (except when prohibitive from performance point of view)
· If possible, to confirm by visual check from air-craft or control tower that no fire or visible damage exists
· To be very restrictive as regards continuation of flight as damage may not be immediately discovered but may deteriorate and make a continued flight hazardous

AOM text covering malfunctions during takeoff emphasizes that once a decision to reject a takeoff has been made, maximum braking and reverse thrust shall be used to stop the aircraft in the shortest possible distance.
This philosophy is in line with NASA recommen-dations following an investigation into a from close to V1 rejected takeoff accident where more than 900 m excess runway available was used for bringing the aircraft to a gentle stop. (Stopping capability of the aircraft was only partly utilized)

DECISION MAKING. In general, your decision making process requires two kinds of information, current and background and this information must be integrated and acted on in seconds. You should be aware that the need for an RTO can occur on every takeoff and should anticipate the problems that may trigger one. Obviously in any situation the more background information you have, the faster and more accurate the decision making process can be.

There cannot be a rigid set of rules for all occasions, as the variables are too complex for that. You must think through all available information and then, should the remote occasion occur, hopefully you will be able to make the right decision.
CONCLUSION. Clearly the decision to continue or reject the takeoff, for whatever reason must rest with the PiC. Unless the situation which is leading to a go/no-go decision makes it imperative to remain on the ground, the chances of success are better by continuing the takeoff and the then determining the next course of action under less stressful and time critical conditions.

The purpose of this discussion paper is to invite you to think, perhaps more critically, about this very important phase of operation and to suggest that a better understanding of the performance characteristics at and around V1 could be advantageous. Rejected takeoffs are still a formidable contribution to the statistics of aircraft accidents and personal injury. After weighing the various factors there will be times when rejecting the takeoff from speeds very close to V1 is justified but, perhaps, after considering all factors, this will not happen quite as regularly as we have seen in past years.

Think and you will probably attain a higher level of awareness and be better prepared for this crucial, multimillion dollar split-second decision.