=shipbuilding =manufacturing =military
To make a large ship, you buy
parts, assemble them into modules, and then assemble the modules into a
ship. The trend has been all of those becoming bigger.
There are 2
basic approaches to dividing a ship into modules: vertical slices assembled
back-to-front, and flat sections assembled bottom-to-top. For container
ships, vertical slices are better. For cruise ships, bottom to top is
better. The modules are lifted into place with large cranes, and generally
welded together by hand.
If you want to do this cheaply, some
necessary steps are:
1) Buy parts with good cost-performance.
2) Design the modules so
integration is simple.
3) Assemble the parts into modules efficiently.
4) Have the equipment to assemble large modules.
Here's a video of shipbuilding with vertical slices. Here's a video of bottom-to-top construction. Note that's done indoors, which avoids rain interrupting work and possibly damaging components.
Why are US Navy ships expensive? Because all 4 steps are failed at.
- Parts for the US military are expensive.
- The design-for-manufacturing
of US Navy ships is poor.
- The workers in US shipyards aren't set up to
work efficiently.
- US shipyards don't have the proper equipment to build
or assemble large modules.
Here's a timelapse of the USS Gerald Ford carrier in a drydock. Note that the cranes lift relatively small pieces, and that it sits in a drydock for 4 years. Here's a promotional video from Huntington Ingalls, the largest US military shipbuilder. The work setups that they decided to show off are somewhat depressing.
Why would US shipyards fail to make worthwhile investments?
- The US government makes cost-plus contracts because it no longer has the
technical competence to write mostly-fixed-cost contracts with
pre-negotiated adjustments. This reduces incentives for firms to minimize
their costs, and they don't try very hard to do that.
- Unions at those
shipyards are run by and for their most-senior employees. They negotiate
deals that block firms from reducing spending on labor.
- American MBAs
ruin everything.
I'm more interested in the overall decline of US manufacturing and overall inefficiency of US government procurement than in shipbuilding in particular. I think the main generalized problems here are:
A) lack of engineering talent on the government side
B) lack of
investment at companies because of management incentives
(A) is partly the lack of civil service exams due to disparate impact rules, but I mostly blame politicians - so it's really voters' fault, so it's really media's fault, so it's really consumers' fault. Well, if you thought any of the most recent several US presidents were competent or non-corrupt, you're part of the problem. (For example, Obama appointed William Daley as Chief of Staff and yet most Americans don't even know he was a Daley Machine guy.)
Here are some approximate costs of things by mass:
steel: $0.25/lb
container ship: $2/lb empty
cruise ship: $3/lb
Toyota Camry: $7/lb
Burke-class destroyer: $100/lb
Stryker:
$130/lb
787: $220/lb MTOW (negotiated price)
Chinese smartphone:
$500/lb (retail)
It's reasonable for a destroyer to cost 10x as much per empty mass as a
container ship, but that still leaves US Navy ships being ~5x too expensive.
The
Sejong-class destroyers of Korea are basically copies of Burke-class
destroyers at half the cost. Most of the remaining excess cost is presumably
from design inefficiency and overpriced parts.
A nice example of
inefficient design in US Navy ships is the Ford-class carriers using linear
electric motors for some elevators, rather than normal electric traction
elevators. Using linear motors that way is extremely silly. While an
electric aircraft catapult isn't inherently unreasonable, the aircraft
catapult design of those is also fairly silly. You don't want to use only a
small fraction of the electromagnetic elements of a system at a time, and
you don't want to use linear motors for systems that use lots of force at
low speeds or when stationary.
Another dumb ship
design thing the US Navy did was plan to use railguns on the
Zumwalt-class destroyers. Apart from the energy-inefficiency,
cost-inefficiency, and poor durability of railguns, they also make no sense
as ship weapons, because small high-velocity things lose their kinetic
energy to air resistance at the ranges involved in naval warfare. Also, at
such long ranges, projectile guidance is important, which is incompatible
with extremely high accelerations. Of course, those grand railgun plans have
now been cancelled after eventually colliding with reality. The Zumwalts got
cancelled too, and now it's not clear that the US military-industrial
complex is capable of designing and building a functional new ship or tank
at all.
A Burke-class destroyer has about 300 sailors, resulting in a labor cost
comparable to the (high) construction cost. So, a more affordable US Navy
would also require reducing crew needed. The US Navy likes having lots of
people onboard for damage control, but that also means more people
inconvenienced and at least potentially in harm's way. Really, a destroyer
like that should only have maybe 80 sailors.
Japan, China, and South Korea have all heavily subsidized shipbuilding, but net Korean shipbuilding subsidies might have been less than the excess costs paid by the US Navy. Why have they subsidized shipbuilding so much, anyway?
- The same reason venture capitalists have subsidized Uber rides: they want
to win, get a monopoly, and then raise prices once other countries have lost
their shipbuilding capabilities.
- Because the commercial shipyards could
start making military ships in case of war: the subsidies are a form of
military preparedness by maintaining an industrial base, and competing
internationally prevents more straightforward investment in such capability
from simply being wasted.