Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

NTSB Report: Glass cockpits have not led to expected safety improvements

Wikiposts
Search
Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

NTSB Report: Glass cockpits have not led to expected safety improvements

Thread Tools
 
Search this Thread
 
Old 12th Mar 2010, 12:19
  #61 (permalink)  
 
Join Date: Jan 2001
Location: Dallas, TX USA
Posts: 739
Likes: 0
Received 0 Likes on 0 Posts
The 2 basic benefits of glass technology are to decrease workload and increase situational awareness. These benefits were thought to increase safety. Statistically glass has increased safety in airline use, where the accident rates are much lower in the generations of aircraft that have these systems. If ever there was a system that illustrates the axiom "don't work harder, work smarter", glass systems are a perfect example.

But even in airline use (as we've recently seen), a failure to understand the strengths and weaknesses of glass systems when a systems failure occurs, can result in the crash of an otherwise perfectly servicable airplane.

Glass is not perfect because the programmed computer logic within it, is designed and written by fallable human beings. The information and map databases contained within are not always accurate either for the same reasons. Anyone who's ever used an auto GPS device for any length of time knows the internal map database isn't always accurate, and the GPS doesn't always select the best route. However that knowledge doesn't prevent the GPS device from being useful, instead it promotes more intelligent use of the device.

It should be the same with glass equipped aircraft. You have to know the glass system, including it's strenghs and weaknesses. Read the service alerts for your system (Garmin for example has these on their website), so you'll understand recently discovered problems. Don't expect glass to be magic and don't expect it to be infallable. It can't take you though unsafe skys and unsafe situations, that your avaition skills tell you are unsafe. If you learn to use glass properly, do expect it to be useful, but within its limits. Also be prepared to turn some of the automation off, if it's safer to fly without it.

Just wanted to add this - Humans will NEVER be taken out of the loop when it comes to operating aircraft. They may sit on the ground in front a console instead of sitting in a cockpit (as in military UAVs), but they will ALWAYS be required. Even Airbus has never attempted to automate the taxiing of an airliner on the ground. Compared to airpseed holds, climb rates, and heading changes at waypoints, ground taxiing is far too complex and unpredictable to automate. My point? Automation will ALWAYS have its limits, because automation can't think outside of its programmed logic, the way humans can. It took me nearly 3 decades of my professional career working with automation to understand this, mainly because I came up in this business in the golden era of "computers can do anything". No they can't, it's a machine like any other machine, with limits just like any other machine.

Last edited by Flight Safety; 12th Mar 2010 at 12:59. Reason: To add the last paragraph
Flight Safety is offline  
Old 12th Mar 2010, 12:59
  #62 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,451
Likes: 0
Received 9 Likes on 5 Posts
“The introduction of GA glass cockpits has not led to expected safety improvements”; but what was expected.
An associated presentation suggests that the expectation was an overall improvement in safety and not anything specifically. The ‘glass cockpit’ accident rate trends towards IFR flight and with lower experience pilots. Thus the failed expectation may be related to problems with ‘glass’ in providing situation awareness via the displays, or that the pilots lack instrument flying skills either generally, or specifically in glass cockpits.

From a very limited evaluation of a modern ‘wide screen’ GA display, the capability for accurate instrument flight appears less than with a conventional format. Whilst large scale ‘ground’ visualisation provides attitude information it can detract from speed and altitude awareness.
Some displays have been promoted as presenting the big picture instantaneously, but IMHO none are as good as ‘dial’ instruments - ASI, Altimeter, and VSI.
It is the skilled use of these instruments which enable the formulation of the big picture. This is the mental model of the aircraft’s flight path and which is taught and enhanced via an instrument scan.

Many commercial aircraft EFIS suffer similar problems, but the difference in comparison with GA could be greater use of automation, which might alleviate some of the EFIS weaknesses, and higher instrument flying skill levels, although these may suffer derogation with automation use.
safetypee is offline  
Old 12th Mar 2010, 20:03
  #63 (permalink)  
 
Join Date: Jan 2008
Location: The Smaller Antipode
Age: 89
Posts: 31
Received 17 Likes on 10 Posts
I think one of the problems with Glass cockpits is the ability for manufacturers to add too many surplus 'attractions' to sell more units, the old planned obsolescence ploy.

The only requirement is to move a piece of machinery safely from A to B, all the rest is to help avoid hitting another piece of machinery ( or the ground ! ) on the way, all the old aircraft managed that perfectly well, by and large. All the add-ons that were invented like Altitude Alert, TACAS etc. came about as the volume of traffic increased and the ATC environment became more complicated, and is a very valid reason - within limits - but the Glass Cockpit Industry is progressing like the mobile telephone industry, the examples of which now do just about everything except make a simple voice contact between two parties, and are complicated to use as a consequence - at least to me !!

I differ in this opinion only with regard to INS, ( now superceded by GPS )the coming of which,in my opinion, was the greatest advance in aviation in our present lifetime.

I have a friend who was an Electronics Engineer in the aviation inustry, and who now owns his own aeroplane, he has fitted it with a Glass cockpit, including auto-pilot and virtually a full FMS system and I believe he is working on an auto-land facility - but it is a Microlight ( LSA ) ! Microlights are forbidden to fly at night, or in IFR, so this is just for fun, and has no place in the general Microlight World of course.

The points mentioned about G.A./ Recreational pilots not being current on the various displays available, and therefore more at risk, is very valid, I've just been type rated on a new, to me, microlight type that is fitted with a digital RPM readout - I hate it, can't see it easily, can't see a 'trend' out of the corner of my eye as the large needle ( not ) moves. Bit like a digital watch, one looks at the 'picture' to tell the time, and the traditional " T " layout was familiar on all aircraft.

Still, I must confess that youngsters who have a couple of years experience of Flight Simulator who now come along to learn to fly, are a breed apart, I might not send them solo, but I might give them an Instrument Rating on their first lesson! - but Airmanship ? that still has to be gained, and that is the same be it glass or steam.

( I get my own back when we start stalling exercises - when you stall a computer game you don't fall off your chair ! )

back to my cave
ExSp33db1rd is offline  
Old 12th Mar 2010, 20:42
  #64 (permalink)  
 
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes on 0 Posts
exspeedbird (excuse spelling)

I agree with you that airmanship still has to be gained...and like simmering a stew, takes a long time.

Glass can be fine...if the pilot has the time to learn it, understand it and practice it...and has a steam gauge backup that is big enough for old eyes to see easily. but airlines don't want to take the time to teach you...you will learn on the line is often a phrase I have heard.

But airmanship...that takes time...and a wee dram of luck.
protectthehornet is offline  
Old 12th Mar 2010, 21:02
  #65 (permalink)  
Trash du Blanc
 
Join Date: Mar 2001
Location: KBHM
Posts: 1,185
Likes: 0
Received 0 Likes on 0 Posts
This topic is red-hot at my company right now.

Here's what I think: when glass/FMS/two-man cockpits were introduced in the 1990's here, all the pilots were coming over from rope-start aircraft. They had the requisite skills, airmanship and situational awareness to fly analog already, so the glass training was just to learn the new capabilities.

Now, we've got guys/gals coming up that have never flown round dials - they went from RJ's to the 757 to the 777. Not a bad way to come up, but it does require a little more emphasis on basic airmanship that was not previously required.
Huck is offline  
Old 12th Mar 2010, 22:16
  #66 (permalink)  
 
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes on 0 Posts
Starting your career in a glass cockpit must be quite an experience.

Us old guys started with baby steps. Starting with taildraggers and eventually getting into electric airplanes that had batteries to start the engine and a radio plus lights to fly at night. Much much later we got HSI's so you didn't have to fly back course localizer approaches flying opposite the needle. Then one day we got to fly a plane with an autopilot. Wow. It held altitude and heading so you didn't have to do it. The round gauges were now a welcome help after learning to fly with a needle ball and airspeed. Then came autoland and you didn't even need to touch the controls, just monitor them.

Now all of that is bypassed so our new young pilots are taught how to manage a flight guidance system. Our new SOP's at a lot of airlines throughout the world is to encourage total automation.

I hope it works. We will eventually see. Once the Sully generation is gone.
p51guy is offline  
Old 13th Mar 2010, 06:22
  #67 (permalink)  
 
Join Date: Jan 2008
Location: The Smaller Antipode
Age: 89
Posts: 31
Received 17 Likes on 10 Posts
.......but it does require a little more emphasis on basic airmanship that was not previously required
Disagree there, basic airmanship was always a first requirement, but I think it might need a greater 'understanding' of why various bits of information are being presented, and where to look for that information when one needs it, e.g. you need to know speed, height, heading, engine parameters etc., and it was probably easier to assimilate all that when one had direct hands-on control over all the variable factors, rather than just sitting watching the computer do it all and wondering why it's doing that now !!

Starting your career in a glass cockpit must be quite an experience.
Not sure, after all they have nothing to compare it to, and are being taught how to use the tools they will use for the rest of their lives, so will probably build from the bottom up and think nothing of it.

I started to fly with a bunch who had previously flown Tiger Moths, and they were sh*t scared of the prospect of flying the Harvard - our ab initio trainer - whereas I was an Innocent Abroad, it was an aeroplane, wasn't it, and weren't we going to be taught to fly it ? Of course they solo'd sooner than me, they had some basic handling experience - and airmanship - that surprise, surprise translated to handling the Harvard, just another aeroplane.

I recall another Nav. instructor telling me that I'd never make a navigator until I'd been over Berlin with the shells coming throught the cockpit whilst I tried to work out the wind velocity by flying three headings 120 deg. apart and assessing the drift on each leg through the drift sight ! I never had to,but years later I despaired of my own students attempts to use the sextant to navigate across the Atlantic, but they never had to, by the time they were ready to join the line, INS had come along.

Later on, teaching young co-pilots who had experience of Flight Simulator, a colleague remarked that whereas they could fly an instrument approach to minima better than we ever could ( maybe ! ) they had trouble breaking out and connecting the real aeroplane to the real World, whereas the old (but considerably younger than I am now ! ) W.W. II bomber pilots that we started our airline career with couldn't fly an ILS to save their lives ( well, some of them ! ) and yet drop out of cloud too high, too fast, not configured and say "The runway is over there - Sir " they would straighten out and fly an immaculate visual approach and touchdown - even with a tail-dragger.

It's Horses for Courses, each generation have their own mountains to climb.
ExSp33db1rd is offline  
Old 13th Mar 2010, 14:25
  #68 (permalink)  
 
Join Date: Aug 2006
Location: Nairn, Highland
Age: 85
Posts: 159
Likes: 0
Received 0 Likes on 0 Posts
I don’t have much experience of glass cockpits although a hybrid was introduced into our 146s just before I retired.

You people seem saying that the all-singing all-dancing display has serious drawbacks when compared to old fashioned round dials; I can well understand the reasoning. But do you have problems with electronic displays that look and behave like the old analogue instruments? Surely modern electronics are far more reliable – and probably cheaper - than “steam” driven instruments? Does, for example an electronic (digital) turn and slip exist? I have in the back of my mind the concept of an alternative display that mimics traditional instruments but is actually glass (electronic). It would be a good system in training aircraft but being so out of touch now (12th year of retirement), for all I know such things already exist.

Jack
jackharr is offline  
Old 13th Mar 2010, 19:51
  #69 (permalink)  
 
Join Date: Aug 2005
Location: fairly close to the colonial capitol
Age: 55
Posts: 1,693
Likes: 0
Received 0 Likes on 0 Posts
I think the problem the NTSB presents is simple. Replacing numerous needles, dials and readouts with a handful of CRT's (later LCD's) held more promise than it delivered. Having related information collated onto a single display eases roving eyes and the addition of trend tell tales and critical number bracketing will surely enhance the safe operation of any aircraft. Or will it?

I recall reading a human factors paper put out by either Stanford or MIT in the early eighties where they looked at how we interpret single device data (gauge) compared to one item among multiple readouts on glass.

It was discovered that a discrete device generally produced a stronger stimulus in comparison to a number or graphical readout on a CRT screen. The difference had something to do with how our mind works - words to the effect of "more 'weight' was assigned to the stand-alone gauge" and one reason given was "the gauge occupied a physical station on a panel board or aircraft cockpit."

The recommendation? You guessed it. Training.
vapilot2004 is offline  
Old 13th Mar 2010, 20:25
  #70 (permalink)  
 
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes on 0 Posts
centerline thrust multi engine rating

some of you know that you can get a multi engine rating limited to center line thrust...like a cessna 337, citation (some) and even the F4 phantom (according to an old phantom driver)

perhaps the time has come to offer a glass instrument rating and a steam gauge instrument rating...ditto for ATP and types .

when you are coming in to land...you watch your airspeed...but with everything plus the kitchen sink maybe the brain isn't set up to do that right with glass?
protectthehornet is offline  
Old 13th Mar 2010, 21:51
  #71 (permalink)  
 
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes on 0 Posts
"The ultimate result of shielding men from the effects of folly is to fill the world with fools."
Herbert Spencer 1820-1903

barit1 is offline  
Old 13th Mar 2010, 22:16
  #72 (permalink)  
 
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes on 0 Posts
We went to Capetown, South Africa a few years ago and coming home to the US my pilot friend and I along with our wives were invited up to the cockpit in mid flight. It was a 747-400 with a glass cockpit. We had a long chat with the captain, invited him to the Reno Air Races, and he was happy to have somebody to talk to. This was prior to 9/11.

I didn't want to show my ignorance so didn't ask but it took me 15 minutes to find our altitude on the glass cockpit. Maybe the suggestion of a glass cockpit rating is warranted. Seems like overkill but maybe a glass cockpit panel signoff, like being signed off for x-country as a student pilot would be good. I see a lot of my friends being concerned about if they are getting a glass sim or not on their next training. Most have little or no exposure to it.

I remember our 2 day course to qualify for our new B737 300 from the 200 using computer trainers. Two months later you got your first flight and you can guess how that went because the FO had the same training. They said you would learn it on the line. From who?
p51guy is offline  
Old 13th Mar 2010, 22:25
  #73 (permalink)  
 
Join Date: Dec 1999
Location: UK
Posts: 1,608
Likes: 0
Received 0 Likes on 0 Posts
I remember our 2 day course to qualify for our new B737 300 from the 200 using computer trainers. Two months later you got your first flight and you can guess how that went because the FO had the same training. They said you would learn it on the line. From who?
Slightly ludicrous given the complexity/capability of the aircraft (even over the more-glass 757) - that would seem to be FAA-led rather than pervasive over the world though, as other regimes required greater training on transfer.

Not wishing to divert too much, the FAA has abdicated so much on their reg responsibilities as compared to what is done elsewhere, relying on union agreements that are no longer effective post-deregulation, that it is not fit for adequate oversight. Not much near-term change is going to happen from Washington though...

I think this whole thread is slightly missing the point (aside from the fact that the main subject was GA in the report):

- Quirky design is unhelpful if it differs from instinctive knowledge - regardless of a glass or steam-dial flightdeck (particularly for a GA pilot) - the quid pro quo is that familiarity for the 200hr fATPL is better where training is consistent and designed to the type flown on the line...even if they have not experienced the seat of the pants moments as much...
- Gizmos must be Apple-ised if they are to be useful and not distract from aviate/navigate/communicate principles
- PPL piston ratings should possibly have restrictions based upon cockpit complexity, rather than present limitations based solely on mechanical complexity

Many cockpit designs are killers if you are not instinctively used to them - think of Soviet AIs, or complexity of something like the B1B or B52 - none of the below mass-produced cockpits are standard

http://www.aerospaceweb.org/aircraft...b_panel_01.jpg
http://www.flightglobal.com/blogs/le...%28SEND%29.jpg
http://www.genebenson.com/Neat%20Stu..._cockpit_2.jpg
http://scharch.org/Dick_Baer/_RFB%20...17_Cockpit.jpg
http://farm4.static.flickr.com/3441/...3c8e576f90.jpg

There is a difference between acknowledging poor design and being precious about change though...!

Last edited by Re-Heat; 13th Mar 2010 at 22:39.
Re-Heat is offline  
Old 13th Mar 2010, 23:59
  #74 (permalink)  
 
Join Date: Jun 2009
Location: US
Posts: 497
Likes: 0
Received 0 Likes on 0 Posts
The 737 300 was FMC, not glass cockpit. The VNAV and LNAV was what we had never seen before. The FMC was the transition, not the glass cockpit which had not happened yet. I was saying it takes time to adapt to new technology. You don't do a two day course and sometime later go out and know what you are doing.

I guess that is why I have no problem going back to 727 mode and flying the clearance manually until the automation agrees with the way we are going. I know what the clearance says, I don,t have time to reprogram at this point. Aviate, navigate, communicate comes to mind when things turn to s**t.
p51guy is offline  
Old 14th Mar 2010, 00:01
  #75 (permalink)  
 
Join Date: Jun 2000
Location: Australia
Posts: 4,188
Likes: 0
Received 14 Likes on 5 Posts
I have in the back of my mind the concept of an alternative display that mimics traditional instruments but is actually glass (electronic). It would be a good system in training aircraft but being so out of touch now (12th year of retirement), for all I know such things already exist
I first flew a glass cockpit 737-300 in 1990 and had no trouble converting from the analogue 737-200. To me the glass cockpit was nothing more than a pretty looking ADI and HSI. I hardly ever used the MAP mode apart from curiosity because I could constantly fix my position with the usual combination of RMI needles, DME and VOR/ILS course indicators. I suppose you could call it the big picture.

It was around then that some airlines operating the new 737-300 took the option of staying with the standard flight instrument panel of round dials in one screen that was available in EFIS form. I believe that was a good idea because it avoided the danger of fixation on one instrument that had so much glass cockpit information presented on it - the ADI and its peripherals within that instrument such as drum ASI, IVSI, LOC and GS, ground speed and radio altimeter etc.

Instrument scan was required just like the 737-200 and pilots had no problem with rusty scan technique that is now the bane of the current generation of flight crew. But there again, isn't the current thinking by airlines that instrument scan is far less important that the past generation of pilots think - if only because the name of the game is full use of all automatics from lift off to touch down and everything is done for you by the PFM boxes?

I often wonder how many operators retain that glass cockpit concept of standard round dials? Not many, I bet.
Centaurus is offline  
Old 14th Mar 2010, 01:36
  #76 (permalink)  
 
Join Date: Dec 1999
Location: UK
Posts: 1,608
Likes: 0
Received 0 Likes on 0 Posts
The 737 300 was FMC, not glass cockpit. The VNAV and LNAV was what we had never seen before. The FMC was the transition, not the glass cockpit which had not happened yet. I was saying it takes time to adapt to new technology. You don't do a two day course and sometime later go out and know what you are doing.
Yes, I know and I agree with you - certainly training as you described in not fit for purpose without gaining and maintaining regular currency. The -300 did of course have 2 years' additional development over the 757 that was already flying.

I often wonder how many operators retain that glass cockpit concept of standard round dials? Not many, I bet.
Same bizarre step option was available on the NG transition - you could have it delivered with 3/4/5 instrumentation portrayal on the screens or a full glass 777.

Degraded scan is a real problem at all levels of aviation, though I would go so far as to say that most PPL instructors don't place enough emphasis on it (if at all). Hardly helps when flying with an abnormal glass cockpit, and you don't even have a framework by which you regularly monitor the "right" outputs.

Last edited by Re-Heat; 14th Mar 2010 at 01:53.
Re-Heat is offline  
Old 14th Mar 2010, 05:07
  #77 (permalink)  
 
Join Date: Feb 2007
Location: i don't know
Posts: 320
Likes: 0
Received 0 Likes on 0 Posts
Some adapt faster to a new cockpit than others, it's sometimes also a matter of self percetion ...

I find it however quite revealing that for their direct recruitment on new equipment, airlines make a marked distinction between conventional and glass cockpit experience. Now where do you think this might originate? Could it be statistics of failure or higher training needs? I do think so.

Considering that, it must come as no surprise what the NTSB found out. The remedy is training, which implies cost and time ... and those two seem to be the biggest hurdles in privat and commercial aviation.
GMDS is offline  
Old 14th Mar 2010, 07:48
  #78 (permalink)  
 
Join Date: Jan 2009
Location: alameda
Posts: 1,053
Likes: 0
Received 0 Likes on 0 Posts
when our airline went from low tech (DC9, 737-200, 300, 757) to high tech airbus. it turned out that those with no FMC experience learned the airbus FMC fsater than those who had to forget the boeing way and learn the airbus way.
protectthehornet is offline  
Old 14th Mar 2010, 10:16
  #79 (permalink)  
 
Join Date: Mar 2000
Location: Arizona USA
Posts: 8,571
Likes: 0
Received 0 Likes on 0 Posts
You are 411A and I claim my 5 dollars...
Sorry, Cat, you lose.

In my view, the new glass panels actually has improved safety in GA airplanes, especially in some cases of nightime situational awareness....at least that is what my friends tell me that have 'converted'.

Now, having said this, would I convert my forty year old twin to glass...not likely, I like it just the way it is, with its forty year old airline radios (King Gold Crown) and all.

MY idea of a good glass cockpit is one with nice big windows
Likewise...and I've flown one for the last thirty years...TriStar.
411A is offline  
Old 15th Mar 2010, 09:30
  #80 (permalink)  
 
Join Date: Nov 2007
Location: Whereever I lay my hat
Posts: 49
Likes: 0
Received 0 Likes on 0 Posts
Excuse me?

p51guy:
I didn't want to show my ignorance so didn't ask but it took me 15 minutes to find our altitude on the glass cockpit.
I find that very hard to believe and that is putting it mildly! 15 minutes ? And how many pilots have misread the traditional 3 pointer altimeter in a critical situation (with led to the introduction of the easier to read drum type altitude indicator on some types)?
File:3-Pointer Altimeter.svg - Wikipedia, the free encyclopedia

A look at the PFD of the 747-400:

Photos: Boeing 747-428F/ER/SCD Aircraft Pictures | Airliners.net
KiloMikePapa is offline  


Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.