It depends on what you call 'mass'. I think most people would agree Liberty ships were mass-produced (fairly large production run, assembly line, interchangeable 'parts' welded together)

Even today, you still get production runs of fairly large ships, such as http://telstarlogistics.typepad.com/telstarlogistics/2011/01.... It has two companies each build a series of 10 ships.

I thought even the Liberty ships were basically craft production, not assembly-line. But I admit I haven't studied the story in any detail.

I don't think producing ten items of any design qualifies as "mass production". Even at complexity levels like ten wooden barrels, at ten items, you're still trying to figure out what the causes of variation in your products are. You are a long way from deskilling the individual tasks and doing time-and-motion analysis on them (you probably haven't even identified them yet) and you can only afford to produce the most rudimentary specialized tooling for the product.

So I don't think ten ships constitutes "mass production".

Wikipedia suggests that the Venetian Arsenal might have been "mass production" when it was producing one ship per day --- but that was with 16000 people, so that's still 45 man-years per ship. Compare to Edison's ore processing plant (2000 tons of ore per day per employee, as compared to craft production with 5 tons of ore per day, a factor of 400 times) or the printing press (a Bible now costs two hours of unskilled labor rather than a year of skilled labor, a factor of a few thousand) or Adam Smith's celebrated pin factory (48000 pins per day by ten men, instead of one to twenty pins per day by one man, a factor of 2400 or more).

If we could really mass-produce ships, then we should expect their cost to fall by a factor of 100 or more from the cost of craft production. Everybody could have a yacht. So far, though, mass production of ships seems to be as elusive as mass production of theorems, software, or happy marriages.

Coming back to the 3-D printing idea, printing a newsletter on a laser printer still costs about ten times as much as mass-producing it with a printing press. But only about ten times as much, not a few thousand times as much. You probably could 3-D-print workable ships out of fused quartz, although it will take a lot of work to get the necessary resilience out of such a brittle material. You could surely do it out of fiber-reinforced concrete, as ferrocrete ships have been sailing the seas for decades.