USS Mccain accident report
- Thread starter masterofnone
- Start date
davidej
Well-Known Member
Sounds like typical software to me.
Geeks sit in glass offices thinking up all sorts of bells and whistles that their shiny kit will do.
No need to consult the users if all these are actually wanted or explain to them how they work (if that is the right word).
Geeks sit in glass offices thinking up all sorts of bells and whistles that their shiny kit will do.
No need to consult the users if all these are actually wanted or explain to them how they work (if that is the right word).
jrudge
Well-Known Member
Interesting read ( I am not in the uk so please don’t think I have better things to be doing at midnight on nye !)
what is staggering is that the failures and uncertainties were as basic as left right front back. There were other issues but fundamentally that was it.
on a plane you transfer control - you have control. I have control. The navy don’t seem to do that that who is controlling what was hidden in small type on a screen.
form over function.
what is staggering is that the failures and uncertainties were as basic as left right front back. There were other issues but fundamentally that was it.
on a plane you transfer control - you have control. I have control. The navy don’t seem to do that that who is controlling what was hidden in small type on a screen.
form over function.
Greg2
Well-Known Member
Quite unbelievable! Having stood watches on the bridge of RN warships in the dim and distant past I am at a loss to understand how anyone thinks it a good idea to have that level of automation to do the basics of controlling steering and propulsion on a warship. Too much to go wrong (as it did) and in the event of engagement with hostile forces too much of a risk. Helming a ship is a skill that needs to be developed and in my time using autopilot was rarely permitted so those skills were in abundant supply as a consequence of many hours of practice. Command and control on the bridge of an RN warship is also very precise with absolute clarity of command in my experience.
Unfortunately the organisational failings surrounding the development of the system and what appears to be scapegoating of those at the sharp end is not unusual.
Unfortunately the organisational failings surrounding the development of the system and what appears to be scapegoating of those at the sharp end is not unusual.
Last edited:
Paulg25
Well-Known Member
As an electronics and software engineer I find these incidents staggering but believable. On the few 'large' projects I have worked on (mainly automotive) the one thing that has always struck me was (what I believe) is the unnecessary complexity of the software. One car audio system I worked on had a team of over 30 people writing software and when I left had approaching a million lines of code.
I have a £100 chinese drone which has wifi and GPS. It's great. Knows where it took off from so can automatically come back. Will follow my mobile phone. Will go to way points I tell it to. Never fails. Not saying this is suitable for a warship but an example of what can be done without spending $500,000,000.
I have a £100 chinese drone which has wifi and GPS. It's great. Knows where it took off from so can automatically come back. Will follow my mobile phone. Will go to way points I tell it to. Never fails. Not saying this is suitable for a warship but an example of what can be done without spending $500,000,000.
requiem
Well-Known Member
Unfortunately enterprise software often seems to be built to a "lowest bidder" standard with little thought to the holistic user experience. The NTSB report references an ASTM standard for marine systems design and how the IBNS system design diverged from those principles. Two in particular are that the system should minimise human error under stress conditions, and that it should avoid unnecessary interactions (i.e. too many clicks). The technical issues reported with the system appear to be mainly around edge cases, e.g. AIS falling over when tracking too many contacts. I don't find this too surprising, particularly if the requirements are vague or testing insufficiently thorough.
However, core functionality tends to work, at least in an "as designed" capacity. The errors that led to the collision should not have happened with properly trained and attentive users. I might not expect this with the average person who decides to add a chartplotter to their boat, but I do expect it of the Navy. From photos and images the system does clearly indicate which station has control of each function. It's not really hidden in "small font" either, but on the other hand it could have been implemented better. E.g. overlaying the thrust control with a warning such as "thrust control at lee helm" when inactive would help make that obvious in stressful situations.
I suspect many people on this forum would be able to operate the basic steering and thrust controls, and probably transfer them between stations, without any training at all. Most critically, as the NTSB report mentioned, for each of the mistakes a simple glance at the screens would have revealed the issue had anyone bothered to look.
Criticism can be made of the software, and rightly so, but I think it was right for the NTSB to put the fault primarily on the Navy's lack of effective oversight, with the bridge team's actions contributing.
However, core functionality tends to work, at least in an "as designed" capacity. The errors that led to the collision should not have happened with properly trained and attentive users. I might not expect this with the average person who decides to add a chartplotter to their boat, but I do expect it of the Navy. From photos and images the system does clearly indicate which station has control of each function. It's not really hidden in "small font" either, but on the other hand it could have been implemented better. E.g. overlaying the thrust control with a warning such as "thrust control at lee helm" when inactive would help make that obvious in stressful situations.
I suspect many people on this forum would be able to operate the basic steering and thrust controls, and probably transfer them between stations, without any training at all. Most critically, as the NTSB report mentioned, for each of the mistakes a simple glance at the screens would have revealed the issue had anyone bothered to look.
Criticism can be made of the software, and rightly so, but I think it was right for the NTSB to put the fault primarily on the Navy's lack of effective oversight, with the bridge team's actions contributing.
Last edited:
jrudge
Well-Known Member
What drone please ?
Paulg25
Well-Known Member
SJRC S70W
Make sure it's the 1080 one.
Make sure it's the 1080 one.
chewi
Well-Known Member
I stopped reading the report when I got to "knots per hour."
Nobody who knows marine affairs would write that.
Nobody who knows marine affairs would write that.
requiem
Well-Known Member
The gCaptain folk really should know better. Fortunately the ProPublica article and the actual NTSB report avoid that error, and are more enlightening anyway.
NormanB
Well-Known Member
I have read the report and I am astonished the USN specified such a system. Even if it was flawless it is just so wrong and on so many levels It totally smacks of a lack of understanding of the user requirement and total seduction of applying technology to a problem that just does not exist.
I view the claim that it reduces manpower on a warship with scepticism and for a USN warship even more so. One person dedicated to helm and power, even at Special Sea Duty is well within the capability of a junior sailor and only basic electro mechanical servo type systems are required which would also facilitate auto helm functions. Maintaining a surface plot while on helm duty is no business of a junior sailor and is a Human Factors SNAFU waiting time happen.
There are much better areas of a warship to invest in order to reduce manpower.
I view the claim that it reduces manpower on a warship with scepticism and for a USN warship even more so. One person dedicated to helm and power, even at Special Sea Duty is well within the capability of a junior sailor and only basic electro mechanical servo type systems are required which would also facilitate auto helm functions. Maintaining a surface plot while on helm duty is no business of a junior sailor and is a Human Factors SNAFU waiting time happen.
There are much better areas of a warship to invest in order to reduce manpower.
QBhoy
Well-Known Member
And to think how over staffed the RN and USN are already...you’d think they would have slack to do without such a thing.
Greg2
Well-Known Member
Cant talk about the USN but leaving aside the disproportionally high number of Commodores / Admirals (which the current First Sea Lord is looking to address) I don’t think that the RN couldn be described as over-staffed. Reduced/integrated manning trials were a thing during my service and the complement of modern warships are generally less than the the last generation. Recruitment is a challenge now and everything I read indicates that having sufficient manpower to meet requirements is tricky.
Completely agree that the claims about IBNS reducing manpower are at best questionable but then the numbers cited as being on watch without the system (up to a dozen on the bridge) are dubious unless the USN do things very differently to the RN.
Completely agree that the claims about IBNS reducing manpower are at best questionable but then the numbers cited as being on watch without the system (up to a dozen on the bridge) are dubious unless the USN do things very differently to the RN.
Last edited:
ryanroberts
Well-Known Member
User interface work tends to get out of hand complexity wise very quickly. A mate of mine works on compiler tech for Microsoft and those teams are never much bigger than 5 engineers.
[2068]
...
Typically how many people would be on the bridge of a similar RN vessel?
I thought there would have been some sort of Navigation Officer up there as well.
QBhoy
Well-Known Member
Was always staggered around the numbers it takes in the RN to run an engine room...being an engineer officer in the MN previously. Also remember being staggered that in some or most occasions the RN would contract in service engineers to carry out repairs and servicing, despite such numbers !
Greg2
Well-Known Member
A normal watch on the Bridge at sea would be an Officer of the Watch (command/navigation), a Quartermaster and Bosun's Mate (helm and power controls - one person doing both so rotating betwee them, ships log keeping, monitoring VHF, tannoy announcements, 'shakes' for watchkeepers during the night and any other things that need doing) and usually two lookouts (one on each bridge wing). There would also be support with radar plotters in the ops room. I cant see how the IBNS would reduce this - the OOW and QM/BM would still be required and unlike some MN ships, the RN always had lookouts and would never rely soley upon electronics - nothing like the Mk 1 eyeball!
Levels would increase when special sea dutymen are closed up i.e. when leaving / entering port, undertaking a replenshment at sea, transiting risky areas (incuding very busy shipping lanes) but even then I don't recall anything like a dozen on the bridge.
I think comparing the RN and MN is like comparing apples with pears. The MN run with very small numbers of crew i.e. the minimum required to get from A to B and load/unload cargo wheras the RN operate warships so have sufficient crew to operate and fight the ship 24/7 if neccesary. Can't comment on contracting in service engineers - not my part of ship
QBhoy
Well-Known Member
I was only referring to the engine room. Understand the rest being manned up (sort of, but that’s another story). merchant navy would umbrella a very wide range of applications remember. Cruise ships etc etc. With plenty more people onboard than any warship. Still a fraction of the crew on deck and engineering. Dsvs where there are literally lives on the end of a line subsea. Still only a few engineers on watch.A normal watch on the Bridge at sea would be an Officer of the Watch (command/navigation), a Quartermaster and Bosun's Mate (helm and power controls - one person doing both so rotating betwee them, ships log keeping, monitoring VHF, tannoy announcements, 'shakes' for watchkeepers during the night and any other things that need doing) and usually two lookouts (one on each bridge wing). There would also be support with radar plotters in the ops room. I cant see how the IBNS would reduce this - the OOW and QM/BM would still be required and unlike some MN ships, the RN always had lookouts and would never rely soley upon electronics - nothing like the Mk 1 eyeball!
Levels would increase when special sea dutymen are closed up i.e. when leaving / entering port, undertaking a replenshment at sea, transiting risky areas (incuding very busy shipping lanes) but even then I don't recall anything like a dozen on the bridge.
I think comparing the RN and MN is like comparing apples with pears. The MN run with very small numbers of crew i.e. the minimum required to get from A to B and load/unload cargo wheras the RN operate warships so have sufficient crew to operate and fight the ship 24/7 if neccesary. Can't comment on contracting in service engineers - not my part of ship
Greg2
Well-Known Member
I take the point re range of applications and guess I was referring to containers / tankers - we used to marvel at how small MN crews were and were envious of their notably superior living conditions!
As I say, engineering wasn’t my world but looking at deck crew it is relevant to note that RN ratings are usually multi-skilled so keeping watches and handling lines may not need the numbers but their other specialism will and operating / fighting the ship 24/7 is resource intensive. As an example, down south in ‘82 and ‘83 we were in defence watches for months on end so around 50% of the crew closed up on weapons, ops room, engine control room etc, etc at all times, working watch on watch off. Under this regime the ship could fight immediately without going to ‘action stations’ from scratch. I guess on the engineering side it isn’t just about keeping things running it is also having the capacity to deal with breakdowns, firefighting, damage control etc in quick time to keep the ship operational - not something that the MN have to factor in during peacetime.
Interestingly, I was reading an article about the next generation of frigates and it seems that greater automation and less people is a factor and it was clear to me that my experience is now very dated!
Last edited:
Old Harry
Well-Known Member
What ever would happen in a war situation, sheer madness to rely on "fly by wire"
