Team Vestas grounded

[bclovisp]

...... and quickly recognized that the highest-level crews operate as a machine, greater than the sum of its parts. The smaller the crew for the size of the boat, the truer this is. When you start looking at what that level of racing entails, you realize that very few of the conventions from a tall ship apply, and the traditional hierarchy gives way to highly specialized skillsets.

....

I am saying that the skipper still bears ultimate responsibility and the navigator gets the blame, but that the way these guys run their boats is right, and that if I couldn't trust my navigator not to put me into a reef, I'd have hired another navigator.

I do know this: If your way says every watch change means an extra 10 minutes at the nav station for the new watch captain and every course change means the skipper needs to wake up and double check the nav's work, your boat is going to be slower than mine. And in 399,000 out of 400,000 miles (which is what Wouter's done, I think), I'd win.

I wish you had written this much earlier. It is so critical to understanding of the issues. Probably one of the most important posts.

From a safety critical sense we would call these systems "brittle". This is one of life's eternal trade-offs. The more highly optimised things become, the more brittle they are.

With the VOR, they removed much brittleness (both figuratively and literally) from the boats compared to the VO70. But the human systems remains highly optimised and brittle in comparison. Whether the reduction in crew numbers since the last round has made things more brittle is another matter, but it is clear that this level of competition level optimisation of roles is both crucial if you want to win, and inevitably leads to higher risk. This is the nature of sport.

Where I am unconvinced, is that this isn't soluble. I have been critical for quite some time of the hairy chested macho attitude to human factors in these races. Personally I have a deep suspicion that crew performance could be improved, and race performance concomitantly improved if some time was spent looking into these issues. The constant idea that saving a kilo of mass is more important than some small crew comfort is almost certainly misplaced if the added comfort is able to do things like improve the quality of sleep they get. This accident might be the trigger to start being serious about such things.

+1

and one-design would have been a perfect opportunity for that (and no i'm not saying they should sail on clipper race yachts... there must be an in-between that's fast but still allows to read a paper map inside)

 

[Herr Wizard]

For contrast...I just read an interview with another VOR team (ADOR maybe?) and they stated that their navigator reviewed the new areas once Race Control opened the exclusion zone and noted the shoals. He then created a manual exclusion zone 'round the shoals (used the term "electronic fence") in the nav program to prevent future routing options from automatically traversing the shoals and pressed on.

 

[jzk]

interview was ok, but a few more direct questions about the navigational practices on board would have been good

- how often do you sit at the nav station with wouter and go over the WX, the strategy, and the course that wouter is suggesting?

- does anyone else ever sit at the nav station with him to discuss these issues? if so, how often?

- when was the last time before the accident that you looked at the route - either by yourself, or with wouter?

As far as the interview goes the answer to all of the above can be boiled down to "Thats not my job." - Unless your are asking Wouter. In which case the printable answer is "I've made a big mistake."

That view is also reinforced by the other interviews.

A single point of failure has been set up. A failure happened, given the affected area the result was not that unpredictable. There are no signs that oversight, review or any other method to mitigate such a problem was used.

i'm pretty sure nico _would_ say it's his job to discuss strategy with the navigator

i don't think it's the practice on any race boat for the captain or anyone to double check the navigators course for things they could hit

it's just not done that way - at least not as a rule

It is not about checking the navigator's work. It is about knowing where the boat is and what lies ahead of it. This can be done very easily without affecting the race negatively.

 

[Kingstonsail]

oh fuck jzk...not only a cunt but an old woman.... you made your point... now go back to pretending you are a maritime lawyer.

 

[MasterRobin]

This is the best.

So it's o'dark thirty and you awaken from your sleep to take a piss, afterwards you decide to swing into the kitchen to grab a bite... It's your house, you're intimately familiar with the layout so you don't bother turning on lights. As you begin to head for the kitchen your optimal path is altered...you continue on, aware of the risks but you approach tge kitchen from a different angle, as a result you accidentally kick the coffee table you pass every day stubbing your toe with enough force that you're convinced half your foot is gone...you're now stranded on the floor in the middle of the living room, disabled. You call back to sleep hq to report the damage and wait for assistance... Eventually you're awoken again by the dog licking your wounded toes... Upon return to the bedroom you're interviewed by sleep hq..

"I had full confidence I coukd safely navigate the living room ocean and was aware of the furniture along the route.. The nightlight was out but the information I had lead me to believe I would be fine.. Last thing I really remember was thinking, I had about 40-80 cm of room.. Then there was this horrific crunch followed by an abrupt 180° spin and another hard impact..." "The damage to the toe was severe and at this point in time, I'm not sure if I'll be able to complete the trip, currently looking into the options but nothing final as of yet.
I've gotta thank the dog for his professionalism in staying on scene until I was sorted. Thanks"

 

[Francis Vaughan]

and one-design would have been a perfect opportunity for that (and no i'm not saying they should sail on clipper race yachts... there must be an in-between that's fast but still allows to read a paper map inside)

To be clear - I'm saying that paying attention to human factors and maybe even a bit more comfort at the expense of some additional weight will actually make them faster. Any race when the crew needs to start up the watch system means that you need to pay attention to crew fatigue as a factor in your overall race speed.

Something I have thought about idly for a while is that boats in these races are a systems management problem. One of the clear problems were are seeing uncovered is that there is no overarching systems management structure - not in the sense that the skipper is able to know that everything is always good. Only by directly involving themselves in the navigation do skippers gain the knowledge that the process is working. That doesn't scale. The navigator is devolved all responsibility, and yet is almost only focussed on weather and routing with respect to weather and currents. In a more integrated view there should be a total view of the entire boat state - including the state of the crew. Crew fatigue is a risk and a performance issue. I'm sure navigators make some calls on course changes with a general eye on crew fatigue levels, but I'll bet there is no science to it, nor any codified understanding.

This is an aside to the main question, but clearly related. There is a lot of understanding of stressed brittle systems, and how to manage them. The level of naivety in the VOR is exemplified in the way one of the few secrets left within crews in the precise watch system they use. Clearly no team really knows the answer, and they are all flailing around with their own pet ideas.

 

[estarzinger]

I might note that it is a common layman's mistake to believe that having multiple people responsible for or review a process is an effective way to improve process quality. Statistical process control experience shows that it is not. Individual humans are (generally) only 2 sigma machines and groups of humans are not much better (for various psychological reasons - group think and seeing what you expect to see). So the 'multiple responsible/review' is both inefficient and not very effective. The better solution is to improve the fundemental process and tools to reduce the possibility of errors.

 

[kokopelli]

Clean, thanks for that. I know that was not an easy interview for Nico to do. Hats off to him. And for you for asking at least some of the pointed questions. The other obvious ones will be answered when appropriate.

Cheers

 

[Francis Vaughan]

I might note that it is a common layman's mistake to believe that having multiple people responsible for or review a process is an effective way to improve process quality.

Exactly. It isn't a matter of multiple overlapping roles. Indeed that can quickly get you to the point where things fall between the cracks. (The "everybody knew that somebody was responsible" problem.) Process is usually the place to start attacking these things. Good process need not be onerous or intrusive. But it should, as a side effect, give the skipper enough grasp on how well things are running, and warning when things are not.

 

[MR.CLEAN]

The constant idea that saving a kilo of mass is more important than some small crew comfort is almost certainly misplaced if the added comfort is able to do things like improve the quality of sleep they get. This accident might be the trigger to start being serious about such things.

If anyone inside the event really thought this was part of an epidemic of apathy or the first of many similar calamities, I'm sure you'd see major changes. As it is, you won't, because no one is worried that another VOR navigator will hit a reef.

Here's what you will see:

You'll see recommendations for doing what SiFi did as discussed above on ADOR (exclusion zone inside nav software, alarm, and remember that SiFi was the guy who put the Telefonica Blue on a rock so humiliatingly 6 years ago), for having meetings on shore and going over all possible dangers, for chart and nav software companies doing a better job of dealing with zoom and dangerous objects.

 

[MR.CLEAN]

I might note that it is a common layman's mistake to believe that having multiple people responsible for or review a process is an effective way to improve process quality. Statistical process control experience shows that it is not. Individual humans are (generally) only 2 sigma machines and groups of humans are not much better (for various psychological reasons - group think and seeing what you expect to see). So the 'multiple responsible/review' is both inefficient and not very effective. The better solution is to improve the fundemental process and tools to reduce the possibility of errors.

I thought that was obvious, but apparently not to everyone. Some people think an 8-man boat can or should be run like a global service company.

 

[MR.CLEAN]

But it should, as a side effect, give the skipper enough grasp on how well things are running, and warning when things are not.

And here, the skipper was on deck, sailing along at 17 knots with great VMG. Warm night, great sailing with the blokes, easy navigation with nothing but a 40m seamount to worry about for 1000 miles ahead.

How would some complicated management process 'give the skipper' any grasp other than that things are going well?

 

[bclovisp]

This is an aside to the main question, but clearly related. There is a lot of understanding of stressed brittle systems, and how to manage them. The level of naivety in the VOR is exemplified in the way one of the few secrets left within crews in the precise watch system they use. Clearly no team really knows the answer, and they are all flailing around with their own pet ideas.

Agreed, was just elaborating on one subpoint, the rest being clear and in my mind logical.

As another side not, I would be curious to know how commandos (navy seals etc) manage those "stress brittle systems"? how the highest ranked guy is supposed to "manage" highly trained experts?

They are in even smaller teams in even more stressful situations, with lethal risk, and armies tend to create and revisit procedures over years and years of experience.

In my field (strategy consulting), we have not so dissimilar issues to a VOR boat: small teams of highly skilled people with big egos, time pressure, lack of sleep, complex data to analyze and act upon... which leads to similar pbs, the main one being that the project manager didn't allocate his/her time properly by lack of identifying correctly the main risks/rewards, hence focusing priorities on the wrong stuff and missing a failure (eg while the manager is attending a client interview, the trusted consultant is making a mistake in the model... which is the same as helming/trimming while the navigator is looking at weather without the depth map underneath).

There are some procedures to minimize such risks (daily team meetings etc) but it's also a matter of managerial psychology (identifying weak signals that something/somebody may create a pb).

I would be happy to learn of something more reliable!

 

[corkob]

I guess from your point of view the Bus is backing up to go over the Navigator again.

i think it is pretty obvious that is where the tracks belong.

Have you interviewed Wouter verbraak yet?

 

[corkob]

Just before publicly grinding him completely into the ground.

 

[estarzinger]

In my field (strategy consulting), we have not so dissimilar issues to a VOR boat: small teams of highly skilled people with big egos, time pressure, lack of sleep, complex data to analyze and act upon...

That caused a flash back . . . I cut my teeth at McKinsey, decades ago, but for a brief moment you brought it all back with that post.
I have had an opportunity to study some elite military ops . . . . And the primary answer is they (Generally) have the luxury of a level of training ( both functional and mission specific), which you (generally) do not. They also (generally) have well developed process check lists and go/no go parameters. What they do is technically difficult but not intellectually complicated - somewhat different from what you do. They also expect things to go tits up and actively prepare contingency plans. That is very unusual in business, where generally all the focus is on getting "plan A" perfect. I was educated as a game theorist and that always drive me nuts.

 

[SCANAS]

The constant idea that saving a kilo of mass is more important than some small crew comfort is almost certainly misplaced if the added comfort is able to do things like improve the quality of sleep they get. This accident might be the trigger to start being serious about such things.

If anyone inside the event really thought this was part of an epidemic of apathy or the first of many similar calamities, I'm sure you'd see major changes. As it is, you won't, because no one is worried that another VOR navigator will hit a reef.

Here's what you will see:

You'll see recommendations for doing what SiFi did as discussed above on ADOR (exclusion zone inside nav software, alarm, and remember that SiFi was the guy who put the Telefonica Blue on a rock so humiliatingly 6 years ago), for having meetings on shore and going over all possible dangers, for chart and nav software companies doing a better job of dealing with zoom and dangerous objects.

I highly doubt nav software companies give a toss about team Vestas going aground.

 

[Francis Vaughan]

I highly doubt nav software companies give a toss about team Vestas going aground.

I think you will find precisely the opposite. We are not talking about some huge multi-national monolith companies with millions of customers. We are talking about small cottage industry companies run by guys that care and have a long history in ocean racing, know the guys in the VOR, and will have deep feeling about this.

 

[estarzinger]

I highly doubt nav software companies give a toss about team Vestas going aground.

I think you will find precisely the opposite. We are not talking about some huge multi-national monolith companies with millions of customers. We are talking about small cottage industry companies run by guys that care and have a long history in ocean racing, know the guys in the VOR, and will have deep feeling about this.

You are actually both right. The distinction is between the charts and the charting systems.
The charting systems guys (expedition, Adrania, b&g) will be all over this. The chart guys (c-map, navionics, and the enc suppliers) will be less responsive. There is improvement opportunity for both.

 

[corkob]

I highly doubt nav software companies give a toss about team Vestas going aground.

I think you will find precisely the opposite. We are not talking about some huge multi-national monolith companies with millions of customers. We are talking about small cottage industry companies run by guys that care and have a long history in ocean racing, know the guys in the VOR, and will have deep feeling about this.

So in your view is there a possibility the Nav software may be a causative factor. Mmmm, another factor for the "Clean" slam dunk jury to consider before driving forward and back once more. Do you have a view on the skippers obligations on a boat?