Sinking of the US Cargo Vessel El Faro: Illustrated digest [pdf]
ntsb.gov
ntsb.gov
I can easily see how a graphic might be preferred over text, regardless of data currency, and especially during the 95%+ times of smooth sailing, but to see how this situation turned out is troubling.
I'm only conjecturing here; there's no reason to conclude based on the available information that there is a problem here. But it would be a pretty typical situation in an organization that believes software is something you call up a subcontractor to bolt on -- most places are still run like that, although the culture is slowly changing.
The "offloading scope / spec requirements to a PM" trend is a slow motion train wreck.
If you, or at least someone else on your team, can't clearly state who's going to be using the code you write, what problem they're using it to solve, and how they're using your code to solve their problem... you need to hire an engineer who can empathize with customers.
You'll have to pardon me for asking a stupid question, though -- what other purpose than navigation would a maritime weather report/forecast reasonably have? This seems to me at first look as a protection against liability, rather than an actual clarification of the intended usage of the product.
I guess I could imagine a use case such as operations planning, e.g. "will we likely be able to perform work on deck tomorrow?", where the stakes are lower, but I imagine the potential for misunderstanding and incorrect usage of such a service would be dangerously big.
It's for optimization of future routes -- how do I get from point A to point B while keeping some weather parameters in acceptable range, burning less fuel, and getting there on schedule. And yes, trying not to drive right into a storm. Not quite for trying to maneuver out after you have already driven into one.
Of course you can also look at it and figure out of cleaning the deck is a good idea or not.
I probably shouldn't comment further, but NTSB's full report is at https://www.ntsb.gov/investigations/AccidentReports/Reports/... and they did not seem to find it too confusing.
So we had to build the stupid thing to show him it wouldn’t work. But until then he needed this thing to happen and neither physics nor his most capable people were going to convince him otherwise.
Now, clearly this is a control freak we are dealing with, but also someone in authority who doesn’t trust images or shapes in his head. I had never encountered such a clear case of this happening and it opened my eyes to it happening elsewhere.
Some people have low spatial intelligence. You would think becoming a ship captain would cull out a lot of people with that problem but maybe if you’re good enough at other things it doesn’t slow you down that much? Maybe he had to see things to understand them.
1) they want all to "buy into" their proposed changes. 2) they want all to know how important their work is and why they should remain in the budget.
When there's a bad accident on a ship that was more-or-less following procedures, owners and captains of ships really want to know what went wrong and why, so they can avoid it themselves. This is the sort of document that captains will read and discuss with their crew, not all of whom are highly literate in English, so it needs to be accessible and clear. It seems to me that the NTSB did good work here.
> Then, at 0554 , the captain ordered a turn to port to get the wind on the ship’s starboard side, generating a port list. A few minutes later, El Faro lost propulsion
---
> Log books indicated no more than 26 inches of oil in the sump in the months before the accident ( a ). With this oil level, the bellmouth would not take in air (lose suction) with an 18-degree list to starboard ( b ), but would lose suction with an equivalent list to port ( c ).
They had other serious problems, but loosing propulsion in heavy seas is very serious indeed.
The biggest issue with finding yourself inside one of these in a proper storm is that you’ll get thrown about violently (the lifeboat; you’ll be securely strapped into your seat using a five-point harness, or at least you were in the ones I’ve had evac drills in) for long enough to be as violently ill.
I’ve been told by instructors - no guarantee it’s true, that, but imho sounds plausible - that the #1 cause of death in lifeboats or rafts after successfully abandoning ship is dehydration; you sweat profusely, puke repeatedly until your body can’t take it anymore.
Contrast this with a typical plane crash investigation which produces, at worst, a dozen or two recommendations - that tells you a lot. I would have thought the maritime industry would have been better prepared. So sad what the crew was up against.
There's also the broken anemometer, but I'm tempted to discount that, because it was obvious the wind was from port, indicating (in any counter-clockwise-rotating storm, i.e. any cyclone in the northern hemisphere) you are headed right into it. Winds should have been from astern if you were trying to cross in front of it. When you're convinced of something, sometimes you can stare right at contrary evidence and disregard it. But who knows, maybe having an accurate wind reading would've jogged some brain cells.
If the engine was kept running as you say, that might have helped them, but only if they had used it to turn and get the hell out of there. If the captain insists on using it to continue driving straight into a hurricane, it just makes things worse. And then of course there's all the water they took on, limiting their getaway speed so to speak. In a certain way it seems like the point of no return was the rupture of the fire water inlet. No telling what the ship could've withstood if they had secured the load better and secured the hatches better. Unfortunately all you get with this kind of hindsight is all the things that went wrong, not necessarily what would've had to go right.
An excellent and very readable document, by the way.