Glass Cockpits – Easy to be Hard?

March 10, 2010 by Bruce Landsberg

TAAThis week I attended the NTSB’s conference on glass cockpits and whether they improve the safety of  GA aircraft.

ASF conducted our own study on Technologically Advanced Aircraft (TAA) several years ago and reached the conclusions that glass, or technology alone, does not a safe aircraft make. It all depends on how it is used!

NTSB’s study confirmed  several of ASF’s findings:

1. Cross country aircraft are far more likely to be involved in weather related accidents.

2. Cross country aircraft are more likely to be equipped with glass.

3. Weather related accidents are much more likely to be fatal.

4. That there is a steep learning curve to get to the promised land of greater situational awareness via glass. Training and practice are essential.

Bottom line: Glass per se does not necessarily make an aircraft safer.

NTSB recommended specific training but noted that this is not easy to do with the variability of hardware. I naively offer a humble suggestion that ASF made about 15 years ago when there were five (5) IFR GPS navigator manufacturers all vying for market share. The field has thinned considerably. How about standardizing core IFR functionality?

In that way, any pilot trained on one model could manage basic IFR operation in another. I recall that we came up with 5 areas. Be able to fly from A to B using the flight plan function, modify the plan, execute an approach, missed approach and be able to hold. Everything else would be open to whatever the manufacturers wanted to do. Granted, that is a lot but the industry would help the pilot community tremendously by making it easier. And if you haven’t noticed – the community is getting smaller rapidly. One reason we’ve been told is due to complexity.

As is often the case, we try to bend humans to technology rather than designing it to conform to the human. I grant you that that is a tall obstacle with some people but build for the bottom half of the bell-shaped curve and you might be surprised at well the average person does.

What are your thoughts?

Bruce Landsberg
Senior Safety Advisor, Air Safety Institute

ASI Online Safety Courses  |  ASI Safety Quiz  |  Support the AOPA Foundation

  • Finbar Sheehy


    I found the NTSB’s analysis uninformative and self-serving (the NTSB is paid to deal with problems – and lo, there may be problems!).

    The NTSB – and you – have not looked at whether glass cockpits improve safety, ALL OTHER THINGS BEING EQUAL. As you note, glass cockpits are more likely to be used in challenging, hazardous flight situations like cross-country flying in IMC. The relevant question is, given that an hour is to be flown cross-country in IMC, does having a glass cockpit lower the risk? (And, equivalently, given that an hour is to be flown locally in day VFR, does a glass cockpit lower the risk?)

    The steep learning curve issue is a qualitative observation and doesn’t address the question; making technology easier to use is motherhood & apple pie; no-one disagrees with it (although going from there to argue that we should have just one mother and one flavor of apple pie is likely to cause trouble). And it’s likely that the learning curve is much more relevant for rental airplanes – something an accident analysis might also be able to study.

    I realize that properly analyzing the data is hard, and the data to do the proper analysis may not even exist, but if that’s the case let’s just say so. Either glass cockpits do help, and people should know that and use them on cross-country IMC operations as much as possible; or they do not, and we need to let people know that; or they help owners but hurt renters because of the learning curve issues; or they have no effect, which would be an interesting finding worth following up on (to see why not).

    Thanks for the review of the story so far!

  • herb ludgewait

    Amen. I believe you have it just right.. Many of us fly and teach behind a varity of steam guages and many brands of GPS. We can’t afford in time or money to stay current in all. Add in the other glass panel stuff, and it becomes dangerous. With all the good the new technology can do, your standardizing is the perfect and doable answer. thanks

  • Alex Cooper

    Directly opposing to the NTSB’s report, my flight training school utilizes all glass meaning it is mainly young pilots and student pilots and we have specified university courses for weather and avoiding cross country weather issues, and a GPS class specifically tailored to the G1000 most of out C-172s are equipped with. The learning curve IS great but I believe the learning curve is less for university students , especially in the realm of computer technology. Although the NTSB associates more experienced pilots with longer trips in glass cockpits, Liberty University is the opposite and has not had one accident since it’s flight training program began.

  • Bruce Landsberg


    Many thanks for your thoughts. NTSB focused on PFDs which is the easiest part of the deal. Personally, I found the PFD an easy transition. I don’t care much for digital tapes used for airspeed and altitude but one can adapt. The MFD is relatively straight forward and moving maps are wonderful compared to the arcane display of VOR or even HSI.

    The real complexity of the matter is in the GPS navigator. GA is now infected with the FMS virus. These unit have not been optimized for single pilot use nor is there any standardization, as mentioned.

    As for safety – the new units with TAWS and synthetic vision may help. Here is where the measuring situation gets really complex. We have no idea how many pilots almost got into trouble, saw the trouble on their MFD and got out of it. If everyone reported their “almosts” we’d have something to go on.

    Another factor – also hard to measure – is the antilock brake syndrome – I can the car ahead follow more closely because my brakes will save me. It has been written that people have a set point in risk tolerance and when technology improves the risk the human just changes behavior to move the risk back to where it was.

    Great conversation….Thanks for bringing it up.

  • Finbar Sheehy


    Thanks for the clarification; I hadn’t picked up that they were only looking at the effect of PFDs – which I would guess would have little impact on safety. It gets me into guesswork, but I would expect that the big technology-driven enhancements to safety would be in the moving-map navigational displays (whether MFD or just a Garmin x95) and in the much greater deployment of reliable autopilots in recent years. The moving map display, especially when terrain and navigation information (approaches, for example) are on the display, makes a night-and-day difference in situational awareness in instrument conditions (and, for that matter, at night!). And the value of a good autopilot is so high that I have come to consider single-pilot hand-flown IFR to be an emergency procedure only. I agree that synthetic vision is likely to be a big improvement too (and one that will really bring the PFD into its own).

    I am with you on standardization, but cautious too. I carry my own Garmin x95 and a suction-cup mount when I fly rental planes, because even in VFR conditions I prefer to have a GPS I’m confident I know how to operate. So I certainly understand the desire for standards – and I do have an instrument rating, so I am familiar with the value – and the hazards – of GPS-coupled autopilot operations (I’ve even watched with irritation as a GPS-coupled autopilot flew to the IAF and then began a 135-degree right turn to try to align with the approach path – turning straight into the hills in the process, fortunately in VMC).

    One the other hand, I dislike speed and altitude tapes too. I think we need to – and hope we will – migrate to a more sensible display format (more below), and this would become impossible if we standardize prematurely. Innovations like synthetic vision might have been ruled out, for example, if a premature standard happened to call for a clean, straight horizon (once you define the box, it gets harder to think outside it).

    I take your point, too, about safety being traded off for utility. Indeed, some of the cross-country IMC use of glass technology surely falls in this category. There’s nothing wrong with the tradeoff, of course; better technology should increase utility, and if pilots perceive safety as being acceptable already, they should use most of the new capability to add more utility rather than adding more safety. That complicates the measurement of the effect of technology, no question!

    An aside about those tapes. The altitude tape is practically useless, as far as I can tell; the only useful part is the actual digital readout of altitude. Might as well just put in there in a box and declutter all the space currently occupied by the altitude tape (keep the VSI though). Heck, you could even display pressure and GPS altitude, MSL and AGL, in a small area (that’s 4 numbers; I’d highlight pressure altitude MSL and GPS altitude AGL). The airspeed tape is also useless. Groundspeed, IAS and TAS all need to be displayed somewhere, but digital readings of each are fine. The tape could – and should – be re-purposed as an AOA display, with a moving and very visible indicator that slides up and down the tape to indicate AOA. The only time you need to pay close attention to IAS is when close to the stall, and what really matters is AOA, not IAS (the target IAS will in any case depend on aircraft loading). We all know this; why are we having digital technology display proxy information instead of what we actually need to know? And all of us old round-gauge-readers know that needle position is a valuable visual cue; a moving tape isn’t. So move the pointer, and leave the “tape” still. Also, when the AOA is high (is below 5 degrees, say, and flaps up) you could remove the whole AOA tape and declutter the screen – which adds the value that when the AOA tape pops up, it’s a big HELLO! to the pilot to start paying attention to AOA.

    Aside ends.


    – Finbar

  • Dan Vigesaa

    Manufacturers have some culpability. When I picked up a brand new twin engine turboprop aircraft from the factory some years ago the factory sponsored training facility had no idea how the electronics worked. I complained. I told them that I needed to know how to use the glass. My answer was “in six months you will teach us how to use that equipment.”

    Years ago pilot training was done by the avionics shops. The technician showed the new owner what functions his equipment had and how to use it. That worked when you were installing a GNS 80 or a Genave Alpha 200. That can’t work now when new equipment may have hundreds of separate functions, options and menu items.
    Manufacturers have not made information or training easy for the flight instructors to get. It’s expensive and may require a costly investment in time and venue. The technicians may not have the time, the motivation or the intimate knowledge required to communicate all of the required information to the new owner.
    The result is that new glass owners are encouraged to spend weeks or months and many, many flight hours with their heads down and locked as they try to uncover the mysteries hidden in the depths of their unfamiliar cockpit suites.
    This also means that many new glass owners never tap the depths of the capability of their new equipment. Some pilots I have trained have never used the “flight plan” function but hit the “direct to” mode every time they fly. The very first time they look for their “missed approach” mode may be when they are below minimums and need to execute a real world missed approach.

    Manufacturers need to acknowledge the role of flight instructors and work to ensure that there is a complete and effective support network for this new glass universe we are entering into.

  • Finbar Sheehy

    Obviously I meant to say “when the AOA is low” in the last sentence there…

  • Rex Cox

    I reckon glass stuff is ok,has anyone had problems readin whats on glass,focusing on the glass,I have 3 diff. powers on my glasses..tryin to read n diff planes diff distance then other..from where I sit.,,glass is ok if numbers were slightly bigger…can they adjust numbers to b bigger or smaller on glass to read em better..always like old PFD I can read them got big numbers n letters…….

  • Robert Hadow

    Yeah Finbar —

    Your analysis of tapes was right on.

    I agree with the NTSB that more training is good… but how can a flight instructor disagree with that?

    I’ve got a real beef with the analysis on which the NTSB built its conclusion. Association is not causality. Longer trips into IMC, without an instructor? Sure, you will have more fatals. Just because an electronic primary flight display was on board doesn’t mean it was the cause. Loren Groff described the reliability of his conclusion as p=0.004. There are four chances in a thousand that his conclusion is a statistical fluke.

    Then NTSB member Sumwalt asked the key question: If the weather were the same, and the pilots were the same, and the mission was the same, how would that change the conclusion? Groff was caught completely flat footed. He admitted that he couldn’t answer that. He couldn’t or wouldn’t do that analysis over the last year.

    Sumwalt was insightful and kind. He did not pursue his line of questioning. A college professor would have failed the sophomore who submitted work like this.

    The single valid conclusion one can draw from this study is that the airplane equipped with a glass panel is more likely to come to a nasty end. The converse is also true, a round dial airplane, likely with an instructor on board, flying in VFR conditions within 25 miles of home is less likely to be involved in a fatal accident. THIS STUDY GIVES NO, NO, NO REASON TO BELIEVE THAT YOUR NEXT FLIGHT WILL BE ANY MORE OR LESS SAFE BECAUSE YOU HAVE A G1000.

    For all the tax dollars I pay, I hope the NTSB can spend another year to do the analysis to answer the question that is relevant to you and to me.

  • Teabag

    On a slightly related note, why are we still referring to aircraft designed in the middle of the last century (eg. C-172) as ‘technologically advanced aircraft’. Putting a 200mph speedometer in a 1986 Chevy Cavalier doesn’t make it a Ferrari.

  • Finbar Sheehy

    Very late follow-up: I was struck by the design of this display, which has eliminated the tapes almost entirely and presents pretty much the information I was looking for.

    I certainly can’t claim any originality!