Rolls-Royce is designing drone ships to challenge $375B freight industry | Hacker News Reader

Rolls-Royce is designing drone ships to challenge $375B freight industry

bloomberg.com

137 pointsnoonespecial133 comments

This is interesting because cargo ships are more fuel-efficient at slow speeds [1][2]. But with a crewed ship, slow speed means higher labor costs. I wonder how much remote piloting or auto-piloting of ultra-slow ships can change this equation.

[1] http://people.hofstra.edu/geotrans/eng/ch8en/conc8en/fuel_co...

[2] http://ultraslowships.com/

I find it unlikely labour costs make much of a difference compared to cargo being delayed on slow ships.

I would be more along the lines of, labour free ships don't go on strike/have HR incidents and delay cargo.

Ships actually use very precise algorithms to calculate the relative cost of labor vs. inventory getting delayed, using factors like the depreciation of the cargo, the spot price of fuel, and even changing interest rates in the market.

Oil tankers in the middle of the ocean, for example, will speed up and slow down as spot prices of oil increase or short-term interest rates decrease, since the inventory cost of their cargo has gone up and then down.

That's absolutely fascinating. Do you have a source for further reading on those optimizations? I haven't been able to find anything with a bit of googling.

Unfortunately, most of the papers I found by googling are behind paywalls :(

I for one would still appreciate a link - I think those of us on academic networks should be able to see most papers behind paywalls

Google for "oil tanker algorithms"

This page has a link to an Excel spreadsheet that calculates optimum speed as a function of fuel price and other factors: http://www.intertanko.com/Members-Information/benchmark/Opti...

oil tankers will sometimes even park themselves out in the ocean for weeks in order to not burn fuel while waiting for the projected price of their cargo to be optimal at the time they actually steam into port. Oil tankering optimization is a very complex and interesting problem!

This sounds very interesting! Do you recommend any articles to read on the topic?

The article says otherwise. Labour costs seem rather a significant percentage.

> Crew costs of $3,299 a day account for about 44 percent of total operating expenses for a large container ship, according to Moore Stephens LLP, an industry accountant and consultant.

Operating expenses, not the cost of the ship. Other operating expenses are just fuel, aren't they? Not that fuelling a container ship is cheap.

The capital cost is probably as much as fuel and crew together. Still, if crew is ~20% of costs, it's worth looking at.

$3,299 is EXTREMELY cheap. I don't think that is right. A captain alone can cost anywhere from $600 to $1000 a day. Then you have a Chief Engineer that might just be $100 less a day, other officers, then unlicensed crew. Salary-wise, that number is less than half.

Also, unless they are getting their fuel for about 30 cents a gallon, fuel costs a lot more than $8247.50 (the other 56%) a day.

Correction: The other 56% is $4198.73 a day.

I can only image the frustration of someone lost at sea watching one of these slowly drift right passed them and no way to get on board and no crew to signal for help.

I imagine that some sort of radio beacon receiver can be attached to these drone ships and with enough of them traveling the seas, it might actually increase the chances of finding survivors.

Or even just cover them with cameras on the outside. When something interesting happens, like a ship passing by or a human being waving their arms in the waves, it will alert somebody who can take a closer look and possibly alert the coast guard/navy.

They'll be covered with cameras on the outside, unmanned boats would be way too easy targets for piracy. You wouldn't even have to sneak aboard, just get aboard and find the expensive cargo. Set off a bomb in the engine room and then unload what you can in a reasonably safe time before someone can get close and track you. It could potentially make piracy more viable in currently safe areas.

You'd want cameras to track when anything even remotely in sight is moving unexpectedly. Anything coming close would be reported to the coast guard/navy.

There'll be heavy monitoring on these ships, even just for liability issue in case some dumb guy in a yacht sails right into them and tried to sue.

The "safety" seems to only apply to the drone ship, and not to other craft that will encounter these essentially blind moving vessels. I think that there already enough hazards on the sea without adding drone ships to the mix.

ALL vessels in the future may need AIS transponders.

http://en.wikipedia.org/wiki/Automatic_Identification_System

Do you imagine these cargo ships are crowded with people looking out at the ocean for days on end?

Crowded? No. But there are certainly people looking out at the ocean at all times.

When the ship is not near shore, there's no doubt in my mind that it is safer in the hands of an automated pilot.

It's only when things get crowded and unpredictable -- that is, with lots of non-automated shipping about -- where you will need serious piloting software. I imagine that there will be a long crossover period where ships will set out from a port with a small human crew -- possibly just a pilot and lookouts -- then go autonomously across the ocean to be picked up by another small crew a few hours out from the destination port.

The article mentioned the drone ships will be piloted remotely.

Many larger ports require larger ships to use a pilot who is familiar with the port near the coast. See for example http://www.mpa.gov.sg/sites/port_and_shipping/port/pilotage_..., http://www.pla.co.uk/Pilotage/About-the-PLA-s-Pilotage-Servi..., or http://www.loodswezen.nl/en/home/1593/

That system could stay as-is if the normal crew would be automated away.

You'd still need some sort of pilothouse in that situation, which would tend to negate some of the clear-deck design benefits.

There's also the question of who's going to deploy (and untangle) the boarding ladder that pilots use to get aboard vessels. As well as the prospects of hacking that subsystem in order to divert the vessel.

http://fixyt.com/watch?v=bcSCNFJDy6Y

http://fixyt.com/watch?v=UT05TYaQV8w

Well that's easy you do some sort of authenticated deployment, with a system more secure than the one shown.

Watch those videos.

Now design me an automated deployment system which can handle the operating conditions and environment, and respond, effectively, as good or better than a human can.

Consider that your ladder can get snagged, caught, swing in the wind or seas, collect flotsam or jetsam, etc. Things can get very squirrely at sea.

Sometimes humans aren't the best at a specific function, but through processing and manipulation capabilities, including the ability to use and apply multiple other tools, they're the best means for responding to a general case of circumstances which cover a lot of ground.

I'd say specifically this case doesn't seem very safe to me. For example you could have a rope attach to some rod on the boat, then you could use that rope to attach some climbing device. Of course, the rope has to be lowered slowly and the receptacle must be large so there isn't much to miss.

Not hard to imagine something better than jumping to a ladder on the side of the boat to be honest.

Unless the remote piloting can be handed off to the local expert who could be watching from the dock based control room.

It's already standard for a local pilot to go out to large ships and steer them into port. That would work perfectly in this model.

Yeah. Automate it to within a few miles off shore. Then each port maintains a small staff of captains who are ferried out to each ship and brings it to port.

> Then each port maintains a small staff of captains who are ferried out to each ship and brings it to port.

Which is already the case, so this would be a fairly easy transition.

I am quite convinced that many pilot position exist simply because of local regulation. Unions do control a lot of shipping, especially loading and unloading.

You really think computers of this day and age are less capable? We trust them for a lot more risky stuff than piloting a ship at very low speeds.

I'm kind of surprised we haven't heard more about self-driving trucks. There is a lot more labor to be saved by removing one truck driver from each of the 20 ton trucks that deliver the cargo to the port than a dozen crew from a 10,000 ton container ship.

I imagine Google is on that. There's not a huge difference between a self-driving car and a self-driving truck.

A self driving ship on the open sea sounds like an easier engineering problem than a car/truck, simply because there is more space with less players. Getting into port may be much tougher to solve, although I'd imagine that a properly calibrated computer could compensate better for the action-reaction lag than humans.

Many ports, particularly higher volume ones, have dedicated pilots that will board a ship shortly before it enters the port and navigate it in. This seems viable with a drone ship.

Volvo Trucks is doing a lot of work on that. There are various "tricks" like going along a fixed route, or several automatic trucks following a leader truck with a driver.

So...a train? but more complex and prone to failure?

Arguably so, but freight trains (outside of large countries like the US) have taken a total battering from the truck market.

In the UK for example it's just not worth putting stuff from port -> train -> truck -> store. Much easier to do port -> truck.

Combine that with the EU's very strict laws on driving time and this is a giant market.

When people complain about how insanely sci-fi it feels. I often get the DLR (docklands light railway) in London. It's totally automated - you can sit in the drivers seat if you want. I have no idea how crazy people would of thought of this idea in the 1960s.

PS: I am a huge fan of getting rid of trucks on the road, but in the UK it doesn't make any sense.

Why isn't it worth it in the UK? Is it because the railways are all optimized for passenger travel or just because of the small travel distances?

All of the above, really. It's a crowded country which strenuously objects to building across the countryside, so re-expanding the rail network is hard and the existing network is fairly full of passengers. The key ports don't necessarily have good rail links. Generally once containers are landed they're taken to a distribution center in the Midlands, taken out of the container and re-packed onto pallet trucks to individual stores.

The UK is pretty compact. If you have a distribution center in the middle of it, then just about everywhere you could possible want to get stuff to is less than a 6-7 hours drive away.

And able to go to places (within a few meters, or less) where there are no rails, variable size (when was the last time you had a train deliver a single refrigerator), and flexible to change where to drive and where to without the decade or so it takes to get a new railway and stations planned, approved, build and operationalized.

The Australians have "road trains" which are exactly this.

Trains do that.

It's ludicrous that there are so many long haul trucks on the interstates. The containers ought to be on trains, and trucks used for the 'last mile'. This would save on labor, fuel, maintenance, and roads. (Most road damage is from heavy trucks.)

The reason this doesn't happen is because of the tax and subsidy structure that heavily favors trucking over rail.

> The reason this doesn't happen is because of the tax and subsidy structure that heavily favors trucking over rail.

The reason for the tax and subsidy structure is the Teamsters. So self-driving trucks are how you get back to using rail. First replace the truck drivers with computers, then when there is nobody left to lobby against it, replace the trucks with rail.

Done. Mines in Australia's Pilbara region have had autonomous trucks for some time, and they are expanding nationwide.

http://www.abc.net.au/news/2014-04-25/computer-controlled-tr...

There is a fascinating (if somewhat dry) book about the emergence of shipping containers and the freight industry called 'The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger' I enjoyed it. http://www.amazon.com/The-Box-Shipping-Container-Smaller/dp/...

This sounds like something a company that had never been at sea in the open ocean would announce in order to capitalize on the faddishness of the word "drone".

The crew on freighters is already minimalist and isn't necessary to steer the ship in flat windless seas with no weather. They are there to clean it, keep the engine running, and deal with things that happen because of weather.

Saying you are going to get rid of the crew is like saying you are going to get rid of the guys who do robot maintenance and repair at a highly automated automobile plant.

From TFA: "While the idea of automated ships was first considered decades ago, Rolls-Royce started developing designs last year. Marine accounts for 16 percent of the company’s revenue, data compiled by Bloomberg show."

RR have marine experience.

I believe they just make engines and other machinery. There's a difference between a company that makes a steering wheel and a company that employes a fleet of drivers to drive it's fleet of trucks.

Let me get this straight, you're saying that f'n Rolls Royce doesn't have the engineering chops to assess the viability of future tech developments?

Additionally, you're saying that f'n ROLLS ROYCE builds the equivalent of 'steering wheels' in machinery?

Someone who hops onto a thread in which someone has been dressed down for an utter failure to verify facts from an article with further utter lack of research isn't worth the time.

"Just make engines" seems a little bit dismissive. Their marine division consists of a range of acquisitions which have spent the last 150+ years building pretty much every type of mechanic and electrical component on some of the most advanced ships of their time, as well as complete ship designs.

They may not employ a fleet of ship crew, but given that they keep winning large contracts, presumably companies that do have fleets of ships trust that they know how to handle ship-design and automation.

The article covers many of the predictable objections or challenges: safety, union reception (both sailors and dockworkers, crew costs (up to 44% of containership operations), though harbor pilots would be another), the International Maritime Organization (IMO), regulations, computer hacking, piracy (and ransoming of both crew and cargo).

I see particular challenges being:

• Small craft safety. There are already considerable concerns among pleasure craft and other small-craft sailors for collisions with commercial shipping traffic.

• At-sea mishaps including load shifts and loss of cargo. Containers do fall off ships regularly, as many as 10,000 per year. Hull-strikes with derelict containers are another small-craft concern: http://www.worldshipping.org/industry-issues/safety/Containe...

• Harbor safety. As with autonomous cars, it's one thing to operate in the largely predictable waters of the high seas, another to navigate a complex and highly trafficked harbor or near-shore waters. If crew were required for such passages, the superstructure elimination benefits would be lost.

There's nothing about automation which doesn't preclude augmentation of crewed vehicles for a best-of-both-worlds experience. Crew sizes could possibly be reduced, though not eliminated.

Large cargo vessels typically have a 20+ year life. Don't expect any rapid deployment of this technology, though as with other automation, fixed-course or closed-course applications (canals, closed harbors) might see some initial use.

Did anyone else read this and think that it might be a good way to do away with piracy also? If there's no steering wheel to grab a hold of, there's no way to steal the ship, and with an absence of hostages, it would be a lot easier to deal with attempted theft, too.

And, with a ship unattended, there is nobody to impede your progress as you carve through the side of the ship and start offloading prize cargo.

Pirates usually do not have capacity to offload ship - they use small fishing boats and speedboats that won't hold even one container. Cargo in most containers is not very valuable (who need container of t-shirts 600 miles away from closest shore?). Finding container with valuables (IPhones/etc) is difficult as there is no accessible manifest on the ship and opening containers is time-consuming.

I also assume that boat will automatically inform authorities and navy/security will be able to reach ship in a matter of hours. With no hostages on container ship, security can just blow up fishing boat/speed boat pirates used and then wait until they run out of food/water on a container ship. Pirates won't be able to control engine/rudder so there is not much they can do other than to surrender.

Hmm, but also nothing stopping pirates from looting cargo.

Then again, they could just be floating fortresses. Perhaps if it takes long enough to steal cargo aerial support could intervene.

I didn't, but it could also give way to a new breed of piracy by: - circumventing the auto-pilot's security - boarding the ship and hijacking cargo (no one to stop you)

I'm assuming the ship would be fortified to such a degree that by the time you could "hack" into the steering system or offload any decent amount of cargo, the authorities would have been notified and would be en route.

Not to mention the fact that pirates seem to want to steal the whole ship and appear to be ill-equipped to offload cargo.

Ships already disappear regularly on the high seas.

Some especially attentive news watchers may be aware that a large commercial airliner is though to have been lost without a trace in the recent past.

http://www.cnn.com/2013/02/08/world/asia/asia-piracy/index.h...

Of course there's a way to steal the ship. You do it with a keyboard rather than a gun.

Piloted remotely means there's a way to subvert the remote control. No need to even board the ship until it's safely in harbour.

Large container ships actually go too fast for pirates to catch. It is usually smaller ships that are in danger.

You'd trade one risk for another: piracy via boarding by piracy via hacking, as described in the article.

That might be by exploiting vulnerabilities in the communications and navigation systems, or could be as simple as jamming communications and disabling the vessel by other means. Or hacking the upstream waybill system to direct contents to the port or destination of your preference (already a hacking option with containerized freight). Social engineering (password / token disclosure, bribery) is another option.

Sure, but there are sensors to tell the ship where it is. Give me a GPS jammer and a loop of wire, and I'll drive the ship to my secret treasure island. Of course, there are countermeasures for that, but I think the lack of steering wheel is a particularly weak argument.

Can anyone think of a peace-time disaster at sea that was not as a result of human error? Titanic? Exxon Valdez? Costa Concordia? The 'humans are critical for safety' argument is flawed when we don't know how well robot ships will do.

Sail augmented with solar needs to make a comeback before we run out of oil, it would be cool if ships just found there own way around the globe, safely crew-less.

I don't know anything about the maritime world, but playing Devil's Advocate you're suffering from selection bias: you're naming instances where humans caused the errors but not factoring in the unreported instances of humans saving the cargo, the ship, the personnel or all three.

I think this is a really good point, but it is also necessary to think about software caused errors. It only takes one error to make a disaster!

MS München 1978, large freighter lost to a rouge wave.

SS Edmund Fitzgerald 1975, large freighter lost in a storm

Tamponas II 1981, Passenger ship sinks due to fire, 580 die

Mechanical failure, fire, and weather take many of them. Just a few that come to mind above, but you can read the list too. http://en.wikipedia.org/wiki/List_of_maritime_disasters

So in automated vehicles industry it looks like shipping jobs are the first ones to go. Ships definitely has less navigation complexity than terrain vehicles and longer response times than cars, so deployment and operations of the captain-less ships could be less riskier. If I were working as a crew member its time to jump the shipping!

EDIT: I take my words back. The jobs are not going anywhere because of this : "The International Transport Workers’ Federation, the union representing about 600,000 of the world’s more than 1 million seafarers, is opposed."

Going in a straight line in flat water... yes, easy to automate. However what about hitting waves properly so you don't capsize in stormy high seas? What about that mechanical staff that maintains the engines and keeps everything running properly, because stuff does break down? What about the cost of airlifting a crew to a ship stranded 800mi from shore because the automation failed?

I don't see this happening anytime soon. From a person/ton perspective, large cargo ships are the most efficient form of transit ever created.

Now, how about automating the loading/unloading of these ships that is VERY slow and labor intensive? Oh yeah... we tried that, remember all the west coast port strikes a few years ago?

Controlling an automobile is largely a matter of reaction and responding to circumstances.

Piloting a ship is a matter of strategy. You've got a vessel of hundreds of thousands of tonnes, a half kilometer or more long, with a stopping distance measured in kilometers, to which you can add or reduce energy (velocity) and/or change course only gradually, moving through a fluid which is itself in motion, often with rapidly changing currents changing direction and speed in only a few meters, through waves exceeding 30m in height, frequently operating in total darkness with few if any route markers, or in fog so thick there's no visibility past a few meters -- you cannot see the boundaries of your vessel let alone other ships or navigational hazards.

And that's with modern navigational aids. Consider sailing vessels up through the 1800s which traveled without any artificial light stronger than an oil lantern.

As recently as 1923, seven of a squadron of fourteen US Navy vessels sailed at flank speed into well-known and charted rocks at the north end of the Santa Barbara Channel. In the Honda Point disaster, seven destroyers traveling at 20 knots (37 km/h) ran aground. Two other ships grounded but managed to work free. Twenty-three sailors were killed.

Radio navigation was available and installed but not trusted. Sailing by dead reckoning in a heavy fog, the squadron commander ordered a course change too early and lead his fleet into the rocks.

https://en.wikipedia.org/wiki/Honda_Point_Disaster

Some of the problems you are mentioning are already solved by airline industry. How can an airplane fly on autopilot mode in a heavy fog conditions ? Radars and satellite navigation.

And the disasters you've mentioned seems like most happened due to errors in human judgement. A machine can take lot of information at a very high rate and make decisions that are similar if not better than humans.

Aircraft operational environments tend to include far fewer navigational hazards (though drone proliferation could change this).

Aircraft final-approach zones are very tightly controlled. Any idiot, or even a dumb lump of wood, can float through the water.

Not all that long ago: rowing through a local harbor (a transport mode in which one faces sternwards and periodically checks over the shoulder for hazards) I managed to spot a large timber likely broken from a pier. Perhaps 0.3-0.6 meters across and 7-8 meters long, with an iron spike or bolt some 30 cm long protruding from one end. I passed only 3-4 meters from it. A week or so later another boat in our fleet struck a piece of driftwood and suffered a hull puncture (fortunately just above waterline).

Timber, lumber, telephone poles, and other hazards are frequent in water. They can be significant hazards not only for small craft but for larger vessels with fiberglass, wood, or even steel hulls. It's rare for an aircraft to encounter a tree at 10,000m.

Yes, human error is a significant factor in many disasters, but human response is also frequently a factor in surviving disasters. This is particularly the case where automated or navigational inputs turn out to be erroneous.

The Air France 447 incident is an interesting case study in this: a design flaw (pitot tube) lead to an operational obstruction (pitot-tube icing) resulting in loss of navigational data (airspeed), causing automated navigational systems to disengage (autopilot disconnect) and fly-by-wire systems to reconfigure to alternate-law mode, crew failures in supplying inputs, following loss-of-airspeed-indication procedures, and responding to flight-path deviation, lack of proper crew response to stall, attempted dual control of aircraft by both co-pilot and captain (Marc Dubois). Compounding all of this was the question of what instrumentation and alerts to trust (airspeed, stall, altitude). Something was lying, but what wasn't clear.

https://en.wikipedia.org/wiki/Air_France_Flight_447

And all of that in skies with inclement weather but no specific fixed or floating navigational hazards to worry about.

One more point on the AF 447 incident: sidestick control interface which failed to provide feedback for inputs supplied by the other pilot. Dubois was attempting to correct the aircraft's attitude but was frustrated in this by the copilot's erroneous control inputs. The flight control systems averaged the inputs.

Fail.

So, they're saying it cannot possibly be cost effective, with a global fleet of around 5000 ships, and conservative savings of at least $3000 per day per ship? It cannot be cost effective to develop a product for which the savings would be at least $5 billion per year (of which most can be claimed back in more expensive ships)? Sounds like at least a few people are trying to maintain the status quo already.

$5 billion on $375 billion is a 1.3% savings, with considerable technological, social, political, and other risks.

Seems that there's an argument toward more automation, but a full adoption of the technology to the extent described in the article strikes me as unlikely, and could well have downside costs well in excess of the claimed benefits.

The potential savings are substantially larger. The article cites a 12%-15% reduced fuel use, as well as increased cargo capacity from losing the bridge.

I was simply contextualizing CHY872's claimed savings.

The stated crew savings should account for a sizable chunk at 44% of operating expenses, unless capital expenses are an even larger share. I don't know the numbers and the article doesn't supply them.

Some thoughts:

Crew on ships seems to be of two classes (I know nothing about shipping be warned!)

1) Direct crew: captain, engineer, etc.

2) Support for the direct crew: Ship's doctor, cook etc.

To eliminate #2, you basically have to eliminate #1 (or get #1 so small and with so many free hours in the day that somebody can dual-hat and handle cooking for whoever's left or even for themselves. Getting rid of decent medical care might be tough though.

So let's look at the ship's engineer. Supposedly, there's a reason there's an engineer on board a ship, things break down and need maintenance on a voyage. So automated or no automation, it seems like he's still going to be needed. If he's not, why hasn't that role (and his team) already been eliminated as a labor cost? Not even Star Trek figured out a way to automate away the engineer on a large ship. Smaller vessels don't need an engineer aboard, but their voyage time at sea is shorter, so the "ship's engineer" role is relegated to a land-based job and split across several vessels - we call these guys mechanics when dealing with cars.

Here's a video of a container ship engine room https://www.youtube.com/watch?v=41priD5GJyY It looks pretty automated today, yet there's still a ship's engineer aboard. I bet that the equation is spend a few hundred dollars a day on a guy or risk losing hundreds of millions of dollars of cargo because the automated system had a fault on a fuel pump or something.

But I suppose you could start splitting duty across ships, have an engineer basically helicopter around to ships to do inspections and solve limited issues. But how many ships does an engineer have to fly around to to make up the cost difference of having him flown around vs just having guys on board?

What about the captain? Out at sea a ship can basically go on a kind of autopilot today. https://www.yokogawa.com/ydk/mr/marine/pilot/products/ydkmr-... Better integration with GPS, charts and real-time weather and piracy rerouting data I guess could make blue water sailing a reality. But coastal sailing, docking, berthing, towing, and other complex navigation functions are pretty complex operations. FTA it sounds like they intend a land-based captain to remote control the vessels, but I suppose a system where they navigate to an offshore location and are towed in or a docking captain in flown in to "park" the ship could work. I believe something like this happens with canal transit.

I'm assuming there's other direct crew on board, probably with similar issues. For the support crew, as soon as the numbers of the direct crew go to zero, they all disappear...fast. But suppose they don't then you'll need two 12 hour or three 8 hour shifts of people.

Let's minimize it and say there's a captain, first mate (backup guy), engineer. Then 3 shifts we'll have 9 people. We'll need a cook for each shift, so we're at 12 people. Add in a doctor and a laundry/room (steward) cleaning person and we're at 14 people. It turns out most container ships have between 15-24 crew, so it doesn't seem I'm too far off. https://answers.yahoo.com/question/index?qid=20100924203011A...

The Xin Los Angeles has 19 crew and it's the most heavily automated, largest container ship at the time of launch in 2006. http://www.gizmag.com/go/5853/

Not a container ship, but a bulk goods ship can have dozens of crew (I suppose the cargo handling is more complex) https://en.wikipedia.org/wiki/Bulk_carrier#Crew

   1 - Captain

   Deck Department:
   1 - Chief Officer
   1 - 2nd Officer
   1 - 3rd Officer
   1 - Boatswain
   2-6 - Able Seamen
   0-2 - Ordinary Seamen

   Engines:
   1 - Chief Engineer
   1 - 1st Asst Eng.
   1 - 2nd Asst Eng.
   1-2 - 3rd Asst Engr.
   0-2 - Jr. Eng.
   1-3 - Oiler
   0-3 - Greaser
   1-3 - Entry Level

   Steward:
   1 - Chief Steward
   1 - Cook
   1 - Asst.

With crew costs being such a significant portion of ship cost, I'd guess that automating away these folks as quickly as possible would be in order. Shipping companies aren't stupid and they replace and buy new ships all the time. It seems to me that we're still quite a ways away from eliminating altogether.

I would not be surprised that you can make engine so reliable that it basically never breaks and just need routine maintenance in ports. Also if you have two independent engines and one breaks, you can just use second engine to get to the closest port.

So why haven't they already done that and eliminated a 3rd of the on-board crew requirements?

The expensive part is to have crew (A/C, food, cook, doctor, etc) not to have one more person on the crew.

A ship's power system is more complex than just one engine. Then add on the auxiliary systems (water filtration, sewer treatment, HVAC, etc) on top of that.Yes, much of it is automated, but what happens if a gasket leaks due to poor maintenance and suddenly you have to operate some part of the system manually?

That's why there's someone always on watch in the engine room.

Adding an automated backup to every system would be a considerable cost, and even then a manual option would be required and some-times relied upon.

(Former ship engineer.)

You do not need water filtration, sewer treatment and HVAC on fully automated ship - you can eliminate a lot of stuff that supports humans.

Good point, which seemed to completely escape me. Though many ships still would need auxiliary systems such as non-propulsion power, hydraulics, cargo cooling, inert gas systems, fresh cooling water, etc etc. Point is, it'll still be an immense system that is not so easy to fully and reliably automate.

Not to detract from your general point, but fresh water is used for more than just human consumption.

What about a tele-presence robot? A technician can just operate the robot and fix whatever needs to be fixed remotely.

I can't even replace an o-ring with gloves on, I cannot imagine the frustration of trying to do such a task with a tele-presence robot. I really don't think the technology for something like that is here yet.

I think the tech is already here: http://en.wikipedia.org/wiki/Robotic_surgery. Though I am not sure if it going to be cost effective.

The grippers and tooling on robots like that (for example the "da vinci surgical robot") are way too specialized, and applied only to minimally invasive surgeries. So not something like that more like these robots :

http://www.youtube.com/watch?v=Ek8GKqmJ7n0#t=70

the annoying thing with these robots is that you still have to specialize the hands,grippers, and tools for the task or group of tasks. We don't really have great "really general purpose" robotic manipulators, but we can make really good specialized robots for very well defined tasks, so I don't think the shipping industry is going to pay for an array of expensive special purpose robots. The technology is really not here yet, but getting ever closer, it's the holy grail of robotics and control systems research. Sorry for the ramble.

>Sorry for the ramble.

No ramble at all I found it useful, thanks for sharing it!

The technology employed aboard ships is still tightly coupled with the processes used to maintain it (ie. Human ad-hoc intervention). If you want to automate the process, the first thing you'll think about is how to change that technology to make things easier for automation. Is it actually possible to turn a container ship into lego bricks which can be easily swapped out when broken, and can it be made hot swappable from the ship inside?

Imagine a car which would monitor each of its parts, diagnose an issue on the gearbox, and turn it offline while driving to replace it.

If there is real $ saving the technology will come sooner than later.

(I'm a 3rd Asst Engr. I just upgraded my license to 2nd, but I'm still sailing as a 3rd.)

Maintenance is staggered because there is only so much room for people to work in inside the engine room.

For most ships, it would be cheaper for the company to pay for the crew than lose out on shipping contracts due to being in port for maintenance.

Engineers can detect a lot of problems that machines can't, be it unusual smells, sounds, vibrations, temperatures, etc. Sure there are probes that can detect vibrations and temperatures, but they would have to be placed everywhere (not just the hundred or so usual spots), and then they have to be programmed. I use a handheld device to measure vibrations on running machinery. It's only as good as it's programming, and it's not that good.

Additionally, hoses burst, o-rings leak, filters need changing, etc. We don't have machines today that can fix those problems on the fly.

My last ship I was on was Diesel Electric. It was incredibly automated, yet still we had MAJOR problems because some circuit boards died.

One of my father's friends did maintenance on large bucket wheel excavators in the GDR. He told how they added sensors and electronics in the late 80s until the electronics failed as often as it helped, so humans remained indispensable -- if only to tell false positives from actual defects.

There seemed a clear point of marginal utility in automation.

> "Additionally, hoses burst, o-rings leak, filters need changing, etc. We don't have machines today that can fix those problems on the fly."

Lots of these things could be happening on airplanes but generally don't. Why can't we get airplane-level reliability on ships?

Probably because we aren't willing to pay airplane prices and do maintenance on airplane schedules. I suspect that a container ship usually takes longer to travel between two ports of call than an airplane is legally allowed to go between inspections.

They happen regularly! That's why you have checks between flights, no mater how short the flight. Daily, weekly, monthly and by cycles(flights) or hours maintenance. You have leaks, breakage, electronics get fried, conexions get loose. Aircraft parts are numbered and tracked they hace a limited live span depending the function. I can not imagine an airplane wirking properly and safely without all this care.

Airplanes have pretty low reliability - multiple maintenance man hours per flight hour. But their time in-transit, equivalent to time between ports for a ship, is measured in hours, not weeks.

"What about the captain? Out at sea a ship can basically go on a kind of autopilot today."

There is for sure a certain feel to boating. Although I have never obviously operated a boat of this size I have owned smaller boats and depending on sea conditions there is judgement in play that can't be offset by technology and especially an actual human with seat of the pants feel and judgement. That said it might be possible if the operator was literally in some kind of simulator and able to sense things in a similar way (although you wouldn't have smell or the feel of the air).

This idea though might cut it's teeth in a smaller area perhaps some kind of shuttle between two close ports as opposed to the open ocean or something like a water conveyor by ship.

> there is judgement in play that can't be offset by technology and especially an actual human with seat of the pants feel and judgement.

I thought the same thing at first, but I then I thought that a) These ships are too big and slow to be agile enough to respond to the kind of seat-of-the-pants type navigation I'm familiar with and, and b) I'm sure people said the same things about driverless cars.

I bet Spitfire era fighter pilots thought the same thing, before the advent to inherently unstable computer controlled fighters. And I suspect a bunch of modern fighter pilots think the same about removing the plot altogether. There's a similar significant benefit there too, remove the meat sack from a fighter jet, and all of a sudden 25g turns become acceptable - there's a lot of limitations on a fighter jet based on the requirement to keep a human being alive while "getting the job done".

Is that really true? This kind of "intuition" kills pilots of aircraft all the time. Part of learning to be a pilot is practicing the suppression of these feelings and trusting the equipment.

A smarter idea would be a fleet of automated ships following a flag carrier with the crew in it.

This way, the crew per ship could be reduced, as an engineer and even a captain cost would be divided between the ships. At the very least, the support for the crew would be more efficiently used , even if the direct crew pee ship don't get much reduced.

Moreover, the ships don't need to be of the same class or cargo, or even company: the centralized crew could be responsible for some container ships and grain ships. If needed, hand off between flag carriers could be done in open sea with a follower ship detaching from one fleet and attaching to other, packet network style.

Maybe, but I'm not sure that

a) there's a shipping need to move half a dozen cargo ships at the exact same time between two points

b) the handling capacity of the ports is such that you wouldn't be leaving half your fleet sitting around doing nothing for a few weeks while the other half gets unloaded/reloaded

It's really easy to look at a container ship and say, "autopilot has already been done, lets just apply that." I think the initial bit of getting the boat to drive between waypoints is a pretty easy problem, but once you put this boat out on the open ocean and something goes wrong, you're going to need some very serious fault handling, which means built-in redundancy and consideration for a huge number of possible failures. What happens when you catch RudderNotRespondingException("I tried to turn the rudder to 24.3 degrees and nothing happened.") Most likely you lost feedback, but maybe the rudder actually didn't turn to 24.3 degrees. When does the autopilot realize that it can't make it to the destination. Does the ship stop and wait for a tech to be helicoptered in to plug the rudder feedback cable back in?

Thats one of about 1000000 things that could go wrong that an experienced engineer would have little trouble diagnosing and fixing or working around, that are much more significant challenges for a computer making decisions. Just looking at the complexity of the engine room you can imagine all the possible points of failure, and the number of required contingency behaviors for the autopilot explodes when you start taking 2, 3, or more at a time.

Ultimately a ship is a system which is broadly similar to that of an aircraft. Everything's scaled up, but the basics - a means of propulsion, a control system, a navigation system - are very similar. We can make autonomous systems which are exceedingly reliable (satellites make it to 15, 20 years in orbit without failure or maintenence on a regular basis) and we've been operating large ships for long enough to have a good idea of where maintenence is required.

A ship's engine room may be larger than the engine of an aircraft, but the systems aren't more complicated. Proper maintenence procedures and redundancy where needed are already used in other systems. Why not ships?

The maintenance crew is still needed to keep the engines running. It's just that it can be piloted remotely. This, however, is nothing new technologically speaking. We could very well take away pilots from commercial airliners but it will never happen, or not soon enough. Because a human in control gives a nice illusion of safety.

When it comes to freighters, the cost of a captain or even a crew is negligible. It pales in comparison to the cost of the ship and the goods it carries. Also, insurance companies would not go down well with this.

It's just a question of time before a ship with a human captain will become uninsurable because the automated alternative is deemed safer.

It only seems prudent for a company like Rolls Royce whose marine division includes a substantial amount of ship automation and ship design to invest R&D resources into ensuring they're on the forefront of that development.

Most of the comments here presume that it's unacceptable for a drone ship to break down in the middle of the ocean and be without crew to repair it. What if these ships were designed so that nothing too awful would happen if they floated around in the middle of the ocean for awhile awaiting another ship with a human crew to perform repairs. Or, maybe another drone ship or two to tow a broken one back to land?

My understanding of container shipping is that customers make SLA choices much akin to us Americans choosing between UPS Ground/2nd Day Air/Next Day, etc. UPS uses these varied SLAs to smooth out its use of fleet capacity and for price discrimination. Shippers operate transshipment ports as part of distribution networks much like the hub & spoke designs of the major airlines. These ports have a bunch of shipping containers sitting around awaiting capacity.

Consider the needs of companies that must transport low-value, high weight/bulk cargo. These companies likely already choose the "UPS Ground" equivalent for container shipping. Due to low product value, inventory costs are low (in transit goods are inventory), so it's probably less expensive to have buffers of goods in the supply chain than it is to pay for tight shipping SLAs. Why should these companies care if the variance they experience in shipping duration is due to capacity constraints of manned-ships or that it took an extra two weeks to fix the ship upon which their cargo was in transit?

It very far from certain that a ship without propulsion will just stay afloat for a while. Even big ships can disappear in a hurry if the weather's bad enough.

See for example the sinking of the Derbyshire. http://www.youtube.com/watch?v=9tN4xROtMjI

Url changed from http://www.digitaltrends.com/cars/pirates-eat-heart-rolls-ro..., which points to this. Title edited to make clear that these are designs.

It was a good change. I knew there'd be a better less blog-spammy version but my google-fu failed me tonight. This is one case where the mods are right on the mark and I thank them.

You might have tried the search box at the bottom of this page, it would have found https://news.ycombinator.com/item?id=7297750 and some others that didn't attract many comments.

When a story is good and hasn't had significant attention yet, a small number of reposts is ok. We want good stories to have multiple cracks at the bat.

Make a ship that is automated, give it a crew and wait. Keep fixing things that the computer can't. When you stop doing that, wait. After a decade without incident, remove the crew.

Don't put oil on it until the data proves it's safer.

Aside from maybe lost comms, I don't see much of a difference between captaining a ship locally versus remotely.

With remotely controlled ships, the freight company could use a set of shift workers to control multiple ships at a given time rather than need a team on every ship. I imagine that a fair bit of captaining a ship would be sitting around too, so allocating multiple craft to one captain could work.

Support staff (cook, doctor, etc) wouldn't be required either.

Space devoted to all that could instead go towards cargo.

It's like a parallel universe, captains can go and sail the world during the day and still go to parties in the evening

How would drone ships pollute less than current solutions?

I would be interested to see how they'll deal with pirates.

why design a new ship, just mod an existing ship and done. the floating/propulsion part needs no modifications

pirates are gonna love this.

Only if they want to go where the ship is programmed to go.

Or offload the cargo and be on their way...

Onto what? The inflatable they came in?

Maybe, or maybe the pirates will evolve to use larger ships capable of offloading cargo once these robo-ships become standard.

The thought of a few guys on a small skiff taking over a massive container ship was pretty absurd not too long ago but yet we see it happening more and more due to the financial incentive.

If the cargo stolen is worth it, they'll find a way.

Now THIS is a good idea!

Wait, are you saying that cargo ships aren't unmanned already?

new kind of piracy - cyber-piracy. Hijacking ship without getting up from couch. The Somali pirates will also need to adapt as after getting the ship boarded they would need to find control room and the actual connector to connect to to change the ship's direction, speed ...

Reply on news.ycombinator.com