SEARCH What's New

Human factors in aircraft design


Ever since the earliest aircraft, the pilot was on the mind of the designer. However, as history has shown, it was sometimes the case that little thought was given to the creature comforts of the pilot.

Nor was it always the case that an aircraft was designed for its ease of control. Never-the-less, designers knew that people would be flying their aircraft and in light of that, there were certain considerations to be made. So, we can say with some amount of certainty that "human factors," although not always a completely obvious aspect, have played a part in aircraft design since day one.

Human factors (HFs) has many current day definitions, but I like to think of it in generic terms as the study of how humans accomplish work-related tasks in the context of human-machine system operation. In a sense, machines (aircraft) should be designed to accommodate the limits of the human user (pilot).

But several caveats to this definition are in order. HFs is not ergonomics: e.g. Can I reach the controls? Do I fit in the cockpit? Can I move the control in the desired direction? Likewise, HFs is not biomechanics: e.g. Do I have the strength to move the control? And despite some texts on the subject, HFs is not the correct term to use when talking about aviation physiology issues. So what is a "human factored" aircraft if not these other things?

Simply put, human factored aircraft is one that makes sense to the pilot! The aircraft behaves in a predictable and intuitive manner as afforded by positive static and dynamic stability characteristics. The aircraft’s controls move the aircraft in an intuitive fashion with all control inputs leading to predictable outputs in change of motion.

The aircraft’s instrumentation is intuitive and arranged in meaningful patterns in the cockpit. In a sense, all aspects that support the correct selection of action in the cockpit are ones that lead to a good HFs design.

Pilots will see human factors at work most often on the instrument panel itself. For example, for years now, Piper has placed the tachometer next to the throttle quadrant in light single engine aircraft. This placement, at least for the left-seat pilot, supports the selection of action in the sense that the input device (throttle) is proximal to the output indicator (tachometer).

Likewise, floor mounted manual flap levers adhere to the concept of movement compatibility because they give the pilot a sense of the angle congruent with the actual flap position. This is a powerful indication to the pilot and generally overcomes the apparent inverse relationship of pulling the flap handle UP to put the flaps DOWN.

In aircraft with electric flaps, the shape or angle of the lever and the movement are generally both compatible with flap position but they may be slightly harder to position as accurately as manual flaps without directed attention to the flap lever. In this example, correct selection of action may be more difficult.

One instrument that has created much interest in HFs circles is the attitude indicator (A.I.). As we are all aware, the conventional A.I. is a status display that depicts a stationary aircraft symbol with a moving horizon; compatible with the picture as seen from the pilot’s perspective. We also rely heavily on this instrument during instrument flight.

Older versions of this instrument had a stationary horizon with moving airplane; hence it was congruent with the actual motion of the aircraft relative to a stationary earth. Stranger still is a version were both the horizon and the miniature airplane moved; a display which some research has actually shown to be superior to either of the other above types.

Regardless of what convention was finally settled on, it appears that none of the above is best for facilitating rapid and accurate unusual attitude recovery. In fact, a better indicator for unusual attitude recovery resembles neither the earth nor the aircraft, but a geometric shape that changes in response to the control required to recover the aircraft.

This type of instrument depiction is referred to as a "command" display, similar to the concept of steering bars on a flight director. This new type of command display is not yet onboard aircraft but I suspect that all glass cockpits will have something similar within the next 5-10 years.

A fine example of good HFs practice in action (whether they were aware of it or not) was Beechcraft’s decision to make the early Baron model with the throttle quadrant arranged the same as in the earlier twin Beech aircraft. Clearly, it’s important that when the pilot reaches for the throttle, he or she actually gets the throttle and not some other control. As a result, placement of controls in the cockpit has been standardized.

Unfortunate for Beechcraft, the later standardized throttle quadrant layout is not the same as their early aircraft layout. The problem here should be obvious. If you move from an early Beech model to a later model, or different company model all together, when you instinctively grab for the throttle, you’ll be grabbing the prop (or vice versa).

So we see what started as an attempt at good HFs wound up as a potential avenue for negative transfer between different model twin aircraft.

There are many other not-so-important examples of bad HFs floating around in the general aviation fleet. One example familiar to me is the environmental controls in the Beech Musketeer series of aircraft. Someone, in their intimate wisdom, decided the defrost and heat levers would be "pull out to turn on" and the cabin air (on the opposite side of the throttle quadrant) would be "push in to turn on."

So in the wintertime, several times a month these aircraft would be written-up as heat inoperative! Of course the heat was never inoperative at all. What happened here was that the pilot would think practically about how the levers should work and pull all three out, thereby shutting off the cabin air and, subsequently, all the heat as well!

In the category of selection of action, movement compatibility between similar controls was violated.

The last example of HFs in aviation I’d like to discuss is called the forcing function. With many systems in our aircraft there are opportunities to do things at the wrong time or the wrong way, such as turning the fuel selector to off instead of to the other tank.

One question that immediately comes to my mind is why these controls were designed to allow such mistakes. But in reality, pilots, being the creative creatures that we are, have historically come up with ways of doing things that the best HFs expert could not anticipate. As a result, the best solution is to force the correct action by making the incorrect action more difficult.

Hence fuel selector stops, mixture control stops, safety switches, and the like. Forcing functions in the form of physical lockouts can make an otherwise questionable design safer in lieu of completely redesigning the system. These types of fixes, for what you and I might call a design flaw, are relatively cheap as well.

Some have said that wherever you find humans working with complex systems you can find human factor engineering problems. I have personally found aviation to be full of such problems and thankfully so, or else a lot of my colleagues and me would be out of work!

So until we’ve engineered all those problems out of aviation, you’ll have to rely on good training, some common sense, and the realization that not everything is designed foolproof. Keep an eye out for the bad HFs!

This month’s Pilot Primer is written by Donald Anders Talleur, an Assistant Chief Flight Instructor at the University of Illinois, Institute of Aviation. He holds a joint appointment with the Professional Pilot Division and Human Factors Division. He has been flying since 1984 and in addition to flight instructing since 1990, has worked on numerous research contracts for the FAA, Air Force, Navy, NASA, and Army. He has authored or co-authored over 160 aviation related papers and articles and has an M.S. degree in Engineering Psychology, specializing in Aviation Human Factors.