Checklists & Pilot Awareness

by Harold Green

When I began writing this article, Runway 28 at San Francisco International Airport (KSFO) had just reopened after Asiana Airlines Flight 214 landed short and slow on July 6, 2013, strewing pieces of airplane and pilots’ reputations across the runway.

After sorting out the inane, ignorant comments of the TV reporters, it appears there were four pilots on board. The pundits made much of the fact that the check pilot was on his first flight as a check pilot and the pilot-in-command was making his first landing at SFO in a Boeing 777, never mind that both were very experienced pilots.

These wise reporters were able to ascertain that more training was required and more oversight was mandatory – at least they did this after learning that Runway 28 does not mean that the plane was landing to the east. This immediately brought to mind a few other incidents.

First, an Air France aircraft crashed into the Atlantic while the pilots on board reacted in a totally inappropriate manner to recover from what they obviously did not recognize as a stall. The basic problem was ice blocking of the pitot tubes and they did not recognize that fact. According to the recovered data and voice recorders, the pilots kept trying to pitch the airplane up, indicating they apparently did not recognize the airplane was already stalled.

On a different but related note, a few years ago I had the privilege of meeting and talking with a pilot who had ejected from an F-15 at over 800 mph. It was a night training flight over the Atlantic and while the cockpit display told him all was well, he sensed something wasn’t right. His father had taught him to fly in a light plane and had preached to him to “Listen to the airplane. It’s talking to you.” Even though his cockpit instruments told him all was well, he heard the wind noise increasing and ordered an ejection with only a few seconds to spare before they would have hit the water.

The pilot’s radar observer was killed, but he survived with serous injuries and was picked up by a rescue helicopter as the event had been detected via satellite data link. It was later determined that the problem was a software glitch in the avionics. If he had followed the electronic advice, he would have been dead.

There was also the case of the miracle on the Hudson. A prime example of a situation, which required piloting skills, while perhaps covered in training, could never be fully experienced until it happened and then depended on the pilot being aware of what his airplane was doing for its successful conclusion.

Consider the training regimen that pilots flying in airline and military operations undergo. They are given rigorous flight checks every six months. They are put through simulator exercises that would curl your hair and they must react precisely in accordance with the training manuals. In short, these folks are trained to a very high level. They must know the right procedure for every circumstance that the trainers can conceive of and they must be able to apply those procedures correctly with no delay.

While no training regimen can cover every possible eventuality, these people can hardly be assumed to be under-trained. No one can know how many accidents this training has prevented. Yet, we still have occurrences as what occurred in San Francisco.

In this discussion, I am not attempting to assess the capabilities or the judgment of the pilots because thankfully I wasn’t there, nor have I ever flown anything approaching the performance of those aircraft, so I am NOT qualified to judge the pilots involved. Our goal is simply to look at a possible view of the situation and how it may apply to general aviation aircraft operations.

Recently while talking to a pilot flying for a manufacturer of high-performance general aviation aircraft, he told me his company, in an effort to reduce the accident rate for their airplanes, is initiating a concentrated effort to ensure that instructors train in accordance with procedures stated in their flight operations manual (FOM). A laudable goal, and it may actually reduce the accident rate. However, it has been my experience that flying only by any written fixed procedure is a good way to fly the airplane as long as everything is going as planned. It doesn’t seem to do much when things aren’t going as planned. What’s even more worrisome is that the FOM doesn’t usually give a clue as to how to tell when things aren’t going as planned.

For example, I have had pilots become extremely concerned when the Visual Approach Slope Indicator (VASI) is out. They have apparently either forgotten how to judge height and the touch down point visually, or have not been following the approach so they can adjust upon loss of specific external visual guidance cues.

The point is, training in many cases, emphasizes response to specific situations often not emphasizing the factors that create the situations or the ability to recognize the onset of the problem. It might be that we would be well advised to spend more teaching effort in overall recognition of the conditions of flight. In this light, consider two levels of flight training:  “education” and “training.”

With respect to light aircraft at least, education may be considered to consist of teaching pilots to develop an awareness and recognition of the need to remain aware of their airplanes’ flight profile at all times, and what this profile means in terms of safety. This includes both awareness of the airplane’s flight profile and the relationship to instrument presentations.

Training then becomes the teaching of the proper response to specific situations as in the FOM. This is certainly a necessary part of flying any airplane – more so as the performance and complexity of the airplane increases.

Checklists and formal procedures followed blindly are analogous to the old saw about the lecture mode of teaching: “The best way to get information from the lecturer’s notes to the student’s notes without passing through the mind of either.”

We have all had the experience of skipping an item on a checklist while being sure we had accomplished the task.  Further, as instructors we have had students do the same thing many times. Of course we correct them. However a nagging question remains as to what will happen if a real emergency arises.

What will happen to the checklist item, or the Operations Manual setting then? Ever forgotten to raise the flaps after take off and wonder why the airspeed remains low and the trim is way off?

It generally takes a lot longer to recognize the cause than it should. Remember, the airline cockpit has two pilots so one can check the other. We don’t have that advantage but, with the flow sequence as described in the next paragraph, by using the checklist, combined with awareness of the flight profile of the aircraft, we can achieve at least part of that advantage.

The use of checklists has been improved with the flow pattern approach to checklist use. In this method the pilot performs the necessary steps via a logical sequence to move through the task to be completed. Then the checklist is used to confirm that all required actions have been completed properly. This approach provides the advantage that all items are considered twice and, more importantly, the pilot will review not only the requirements, but his or her own actions as well. In effect this goes partway to adding a second pilot in the cockpit.

Checklist use is only part of the story. A necessary part of a pilot’s education is to instill the need to instinctively and continuously monitor the aircraft’s flight profile. This must also include a thorough understanding of what produces that performance.

Many pilots do not seem to understand the relationship between pitch, power and aircraft performance even though they have had the relationship explained to them many times. Perhaps that is because these relationships have not been reinforced sufficiently during their flight training.

For example, after setting one notch of flap for takeoff, it is not unusual to miss the need to raise the flaps when transitioning to cruise or cruise climb. If the flow pattern approach were used, the pilot would have the advantage of a double check via the checklist. Betcha there aren’t very many people who have not been in the situation of wondering why the airplane is not reaching cruise speed after takeoff, only to discover that the flaps were still in takeoff mode. I confess I’ve done that.

In summary, the points being made are twofold. First, whenever possible, use the checklist not as the primary means of establishing the aircraft configuration, but rather as a necessary means to back up the pilots actions. Second, I believe we need to continuously emphasize the need for the pilot to be aware of and monitor aircraft performance and fight configuration at all times. This should be in addition to, not in lieu of, the FOM and checklists.

EDITOR’S NOTE: Harold Green is a Certified Instrument Flight Instructor at Morey Airplane Company in Middleton, Wisconsin (C29). Email questions or comments to: or call 608-836-1711 (

Share and Enjoy:
  • Print
  • Digg
  • StumbleUpon
  • RSS
  • Facebook
  • MySpace
  • Reddit
This entry was posted in Columns, Columns, December 2013/January 2014, Pilot Proficiency and tagged , , , , , . Bookmark the permalink.

Leave a Reply