Here’s an informational post
on ship construction costs as typified by the Burke class. Data is taken from the Navy’s Shipbuilding
and Conversion budget submission for 2017 (1).
Burke Class Cost Breakdown
Planning $40M 2%
Basic Construction $739M 44%
Change Orders $114M 7%
Electronics $173M
10%
Ordnance $508M
30%
Other $40M 2%
Total $1697M
From the detailed cost
category breakdowns in the document, here are examples of the kinds of items
included in each broad category.
Construction
- Basic hull and superstructure construction
Electronics
- SQQ-89 ASW
- SLQ-32 SEWIP
- IFF/TACAN
- CEC
- Navigation
- SLQ-25 Nixie
- Exterior Communication Systems
- Internal Networks
- STC 3 IVCS
- Main Reduction Gear
- Machinery Control System
- Integrated Bridge Navigation System
Ordnance
- Aegis
- SPY-6 AMDR
- Mk 41 VLS
- Mk 45 5” Gun
- Mk 37 Tomahawk Control System
- CIWS
- SPQ-9B
- Mk 32 Torpedo
- EO System
- Mk 160 Gun Fire Control System
We see that the major cost,
by far, is the basic physical construction of the shell of the ship, meaning
the hull and superstructure. This makes
up 44% of the total shipbuilding cost.
This also gives lie to the oft repeated “truism” that “steel is cheap
and air is free”. In fact, it’s clear
that steel is not cheap and air is not free – not even remotely. All the people who casually call for a little
bigger ship because the added steel is cheap, are wrong. We’ve already discussed this topic (see, “Steel Is Cheap and Air Is Free”) and disproved it and
this data simply adds more support to the recognition that steel is not cheap
and air is not free. Another “truism”
dies!
Hand in hand with the claim
that steel is cheap and air is free, is the claim that electronics make up the
major cost of a ship. As we see,
electronics accounts for only 10% of the ship cost. Aegis, oddly, is grouped under ordnance and costs around $220M. That's around 13% of the ship's cost - not an insignificant amount but nowhere near the major portion of a ship's cost. Another “truism” dies!
As a sidelight, from the
same document, here are the Burke class unit costs for the last several
years. It is interesting to see the very
wide variation in unit cost. There is a
very loose relationship between quantity and cost but I emphasize loose. For example, the highest cost is not
associated with the lowest quantity. Of
course, over that time frame the Navy has been tinkering around with the Flt
III modifications so some variation could be due to that. There were also restarts and partial technology
insertions. My tentative conclusion is
that there likely is a quantity-cost relationship but that other factors
outweigh and obscure the quantity effect.
The Navy’s recent practice of deferring major portions of construction
until after delivery also obscure the true construction costs.
Year Qty.
Unit Cost
2010 1 $2.133B
2011 2 $1.561B
2012 1 $1.904B
2013 3 $1.437B
2014 1 $1.731B
2015 2 $1.497B
2016 2 $2.253B
2017 2 $1.697B
As I said, this is just an
informational post that sheds a bit of light on the cost contributors to the
construction of a ship.
________________________________
(1) Dept of Defense 2017 SCN
Budget, Exhibit P-5c, 2017 DDG-51, FY2017 data column
If I read the list correctly then Aegis/AMDR is part of the "ordnance" and not the "electronics" category, so the "10%" doesn't include these systems.
ReplyDeleteBut this doesn't change the fact that steel isn't cheap.
You're correct. As I reread that section of the post, I was not as clear as I could have been about where the various items that we usually consider electronics were grouped or what their impact is. I've gone ahead and rewritten that part. See if it reads any clearer now.
DeleteSkilled labor isn’t cheap. And Union labor is even more expensive. Then tack overtime onto that as we a facing a shortage of skilled labor especially in some of the fields that affect shipbuilding such as welding and pipe fitters.
ReplyDeleteThe Navy wants cheaper ships? Take some of the millions spent on politically correct programs and put it in vocational training programs in high schools with the open goal of training people for a job in shipbuilding.
My father spent a lifetime building pars for Navy planes. By the time he was ready to retire his shop was doing 60 hour weeks because they couldn’t get new workers. And that was at $30+ an hour.
The President and Sec Ed are pushing a vocational program like you are talking about.
DeleteMM-13B
It doesnt really have much to do with a lack of vocational training. Its more caused by the immobility of the general workforce in this country, the limited local workforce available to shipyards, the higher than average cost of living around shipyards in general (coastal cities), and the volatility of shipbuilding work (kinda tough to agree to move your family for a job that might vanish during the next budget).
DeleteIf the Navy decided to actually invest in rebuilding and maintaining its shipyard facilities and maybe reopened a few around the country to at least be capable of doing overhauls on big ships that would probably go a long way toward stabilizing a more robust workforce. With more shipyards around the country it would also elevate the idea of ship building as possible work in the minds of the general public.
"... stabilizing a more robust workforce."
DeleteVery nice comment although I think you're slightly off on one or two points (like immobility of the general workforce). Still, that's almost nitpicking on my part. I think you've identified some relevant and interesting issues related to shipbuilding, the skilled trades, and vocational training.
I think there's also another aspect to this that ties directly into the naval shipbuilding workforce issue and that is the dearth of commercial shipbuilding in the U.S. We've regulated commercial shipbuilding almost out of existence. If we reversed that trend and could revitalize our commercial shipbuilding it would increase the demand for skilled trades, provide more opportunities in more locations (your mobility point), and, hopefully, encourage more vocational training programs.
Naval and commercial shipbuilding are related on so many levels. We talk about the need to maintain a viable naval shipbuilding industry but we totally neglect the intertwined commercial shipbuilding industry. We need to reverse the regulatory damage done and encourage commercial shipbuilding to come back to the U.S. I'm not saying we return to unsafe, environmentally damaging practices but there has to be a happy medium that doesn't destroy the industry, as we've done. Combine regulatory reform with today's automated and computer assisted design and construction practices and we ought to be able to build a financially competitive ship in the U.S.
What do you think?
Good comment.
Thanks CNO. I agree that a viable commercial ship building industry here would help out immensely (and generally be a good thing) but I dont think regulations can be blamed for much of its demise (hello Jones Act!). Most cruise ships are built in Japan and Europe under more prohibitive regulations and equal workforce pay, they just seem to be dedicated to the industry. A step in the right direction would be massive ship building infrastructure investments (by the government alone, or by a public/private partnership), like I mentioned. No company is going to look at US facilities and think they are a good place to build ships.
DeleteMassive public investment in the rest of the country's infrastructure would also probably help out ship building, making it easier and cheaper to move large quantities of raw materials around the country. Private money will never help out with that since its an enormous cost and an enormously distributed benefit that would help out their competitors just as much as it helped them and make it easier for any new company to take them on in the future.
So yeah, I dont think its regulations, I think its priorities. The US's priorities are not at all aligned with those needed to foster large scale manufacturing, and arent getting any closer any time soon unfortunately.
"I dont think its regulations, I think its priorities."
DeleteI've got to disagree with you on this. I think regulations greatly, and negatively, impacted the shipbuilding industry. However, the impact goes back many, many years. At that time, the U.S. imposed stifling regulations compared to other countries, AT THAT TIME. Also, my blame on regulations extends beyond the shipbuilding industry, itself, to the supporting industries. For example, regulations decimated the steel industry. I witnessed this growing up. As U.S. steel sources evaporated, the cost of shipbuilding went up. Other related regulatory burdens (meaning, added costs) included environmental and OSHA safety regulations - all well intentioned but imposed too fast and too far. All of these regulatory impacts conspired to drive shipbuilding out of the U.S.
Your thoughts about priorities are valid in that Congress and the American people did not (and still do not) make industrial support a legislative priority. That support could have, and should still, take the form of infrastructure, as you mention, and tax reform, medical insurance reform, etc.
Another aspect of priorities is societal. We currently seem to view industrial success as "evil". We're doing everything we can to impose more regulations, shame the wealthy, redistribute their wealth, beat down successful companies, etc. No wonder shipbuilders don't want to set up shop in the U.S.!
Congress dimly grasps that international economies are interrelated but seems to fail to grasp that our own industries are interrelated, as well. Supporting one industry supports all industries. It is the job of government to remove impediments to business success not create more impediments. All right, enough of my anti-government rant!
We appear to be in agreement and just highlighting different aspects of a multi-faceted problem.
There has to be a way of supporting industry without returning to the Robber Baron monopolies of late 19th/early 20th centuries. We are all still paying dearly for their unregulated excesses. Superfund sites dont exist because they have a cool name after all.
DeleteIndustrial success only has a bad name because our primary references to it as a nation are negative. As soon as they stopped hiring private security to murder union organizers they packed up their toys and headed to Asia. The last time industry was great here (post WWII) was at a time of record low inequality, low CEO pay, and sky high tax rates for the highest earners. Making industry great again doesn not have to mean we bend over backwards for the 1% and beg for the pleasure of their table scraps. But enough of my inequality rant!
Our government (congress specifically) is set up to only support local projects with concentrated benefits since congressmen dont lift a finger unless someone is bribing them with campaign donations and local jobs. None of that gets us any closer to supporting a nationwide industrial base.
Distributed benefits are the name of the game but no one (not in the media or politicians) can even acknowledge the phrase exists since it doesnt directly benefit corporate interests. We are willing to pillage money for public schools to build stadiums for billionaires but we cant raise the gas tax one penny to keep bridges from falling down. Priorities.
This is not a political blog so I'll leave you with the last word on this. Good comments. Good contribution.
DeleteYou seem to have an interest so you might consider researching the whole shipbuilding demise and tell us what you find. I'd be interested!
Yeah, sorry, I got a little digressive and soap boxy, thanks for the comments though. The whole Puerto Rico debacle and the Jones Act have refreshed my interest in the industry (along with a recent GAO report about the Navy's maintenance gap which, sidenote, did you have a chance to check out? I would love to hear your opinions about it). I am a terrible writer so I am not sure I will be able to put forth anything worth reading. But hopefully I can keep digging a little bit to at least provide better and more informed comments in the future.
DeleteAnyway here is that GAO link if you are interested:
https://www.gao.gov/products/GAO-17-548
There is a different context of the "steel,air" phrase. That is to avoid complex ship structures when simpler ones will suffice, even if the simpler design will use more steel. Complex designs may use a little less material, but cost much more to build do to increased labor costs. During war time, a complex design will difficult to speed up production rates, and repair of damaged ships will be slow and expensive.
ReplyDeleteMM-13B
Any knowledge of what the Ballistic Missile Defense Agency funds, is it purely R&D plus testing for Burke's Aegis and missiles.
ReplyDeleteSorry, I've never looked into that. Here's a link to the Missile Defense Agency funding that might have some information.
DeleteMDA Budget
With AMDR I'm assuming that this is a Flt III Burke? If so, that's cheaper than I would have thought (Yes, I know, its not the real cost because of the Navy's love of playing budgetary shell games). Still...
ReplyDeleteOne thing that jumped out at me was $114 in change orders. That's not an insignificant part.
I wonder if in WWII, when we were cranking out destroyers, we settled the design and just started building?
That would seem to have two affects on cost, to me. The first is that you eliminate the cost of change orders. The second is that it might bring the numbers/price relationship closer together as you are building to one spec.
It also might make things cheaper in the maintenance end during the life of the ship because the Navy can buy similar parts.
"I wonder if in WWII, when we were cranking out destroyers, we settled the design and just started building?"
DeleteGreat question. My understanding is that, yes, we made constant changes in construction in WWII but NOT during the construction of an individual ship. The changes were made between ships. For a run of 50 Fletcher class destroyers, changes were made from ship to ship but once construction started on an individual ship, significant changes did not occur. That new bridge layout would wait for the next ship, not be wedged into an already under construction ship.
This highlights one of the problems the Navy faces today and that is the very limited number of ships we commit to. If you're building a three-ship Zumwalt class and you come up with a change, you don't really have many more ships to wait on to incorporate the change down the line. You either insert it into a ship already under construction or you probably don't get it at all.
Even the Burkes, of which we've built sixty or so, were purchased in quantities of two or so at a time, over and over, with no guarantee of more. So, at any given moment, there were only one or two ships pending - again, not much chance to defer a change to the next ship. Order quantities in WWII were much larger and changes could be calmly scheduled into a future ship rather than frantically inserted into a ship under construction.
"AMDR ... that's cheaper than I would have thought"
DeleteMe, too. Remember, though, that AMDR is also impacting other costs. The ship had to be redesigned to physically house the radar and cooling and other utilities had to be significantly increased among other impacts. So, AMDR has a direct cost and an indirect cost.
Finally, remember that this is the physically scaled down AMDR that is not the full version that the Navy says it needs.
The irony to me is that in the age of CAD and electronic modeling we were supposed to avoid this type of thing; that we could get the design settled from the get go. That has failed us with the Zumwalt and even the LCS which is supposed to be a mass produced ship.
ReplyDeleteI guess I have this image in my head of a design with computer modeling for a one off test ship that makes us finalize the design. Once the design is finalized we buy no fewer than 10 and make a production line.
Good point on AMDR, I hadn't thought about that, and its impact on the other hull costs.
ReplyDeleteI did know it was the cut down version. Hopefully its better than what we have. Whether it will be worth the cost....
It kind of depends whether the reader understands what they are reading.
ReplyDeleteI'd lay good money that the "Basic construction" line item represents the only the manpower used to build the ship, which isn't just "the hull", but includes all the labour required to prepare, install, set to work and commission the equipment (which includes the electronics and ordnance, machinery etc etc).
A relatively small proportion of this labour is spent on fabricating the steel hull, even less on the propulsion. The main labour extent is on installing what is known as "outfit" - which is essentially all the systems (including electronics and ordnance) equipment within the ship, plus electrical cabling, piping, insulation/fire protection, compartment furniture, paint etc and then setting the systems to work.
What is represented by the Electronics, Hull Mech & Electrical and Ordnance categories is simply the material procured, not the installation labour used to install and commission it aboard the ship.
The $739M cost attributed to basic construction is in fact the labour required across the whole shipbuild, rather than the cost of the hull. As an example, the amount of steel in an AB is of the order of 4500 tons, which at $3000/te reflects $12M or so material cost, from the "hull" total of $81M. Compared to $680+M for ordnance and electronics.
Which, I would suggest, puts a slightly different spin on things.
As a shipbuilder, with experience on half a dozen different builds over the years including naval auxiliaries, frigates and commercial vessels, that's what those budget figures smell like to me.
"I'd lay good money that the "Basic construction" line item represents the only the manpower"
ReplyDeleteYou may be right. The SCN document does not indicate where labor costs are included. Regardless, you're misunderstanding the main point as it relates to the "steel is cheap" concept. That concept is not about the literal actual cost of steel. The concept is that ship "space" is cheap, hugely so, in comparison to all other aspects of ship construction. Thus, the concept claims, there is no significant cost difference between a 300 ft long ship and a 400 ft long one. This concept is used to justify bigger ships across the board, bigger ships to accommodate more weapons and aviation, and hull length extensions for modifications. Thus, if we like Burkes but they're a bit crowded, we should just add 50-100 ft of length to the next Burke because it wouldn't cost anything - the steel is cheap and the air (meaning, internal volume) is free. Of course, nothing could be further from the truth.
Every foot of ship length and every compartment must have the basics of power, ventilation, heating, water, etc., as well as the obvious cost of the steel and its fabrication, lifting, and assembling. It must be fully piped, "conduited", "ducted", etc. in order to be usable. After that, comes the specific use equipment, from some other construction category, like computers, or engines, or whatever. Thus, each foot of length has a significant cost to attain a "basic" level of construction. The data in this post supports that.
And, yes, the cost of labor is part of the "basic" construction cost for every foot of length. Without labor, the steel is just a giant, unformed paperweight, as you well know.
Does that make it clearer?
The data in your post does noes not support your contention that each foot of length has a significant cost, as most competent shipbuilders and naval architects will attest. The only people who use the steel cheap, air free line tend to be naval officers, with limited understanding of where cost lies. What tends to be true is that the marginal cost of making ships slightly larger is cheap, compared to alternatives. While it is undoubtedly true that additional length tends to incur additional outfit cost, it is rarely at the cost of significantly increased system complexity - hence marginal, rather than significant, effect.
DeleteSome of the examples you give don't actually work the way you think. Extra length can actually result in additional speed for the same installed power (or reduced power for a given speed), particularly for higher speed naval vessels, where Mr Froude dominates the resistance at top speed. Increased hull depth can also reduce scantlings required for global strength and hence steelweight (a delicate trade-off). Anyone who has actually designed and built a ship can also tell you that a little extra space internally can actually reduce the labour manhours required to build a given ship.
It isn't a simple relationship, but making the ship bigger (within reason obviously) can in fact get you more capability for approximately the same or marginally higher cost.
The data seems pretty clear and I'll stand by it. If you'd care to present some data supporting your contention, essentially that steel is cheap and air is free, I'll gladly consider it. Lacking any data to the contrary, I'm comfortable that I've made my case!
DeleteThe you posted data does not show how much (if at all) the Burke cost would increase if the hull was lengthened as other anon described.
Delete"does not show how much (if at all) the Burke cost would increase if the hull was lengthened"
DeleteNo, not directly, but it gives a pretty clear idea of what the costs would be. The entire basic construction would apply as well as select other categories depending on the use of the additional space. This is a reasonably extrapolation. It's not rocket science.
Looking at this it does give a good rule of thumb for estimating the costs of some of the ideas discussed on this blog. Adding the planning, construction, hull, electronics (which does not include aegis) and change orders gives us around 1.1 billion for 9400 ton IIa or about $11700 per ton. Yes, I know that isn’t a direct correlation but it does give us a yardstick to guesstimate with.
ReplyDeleteSo the CNO’s 1000 ton Destroyer escort—minus ordinance and electronics is roughly 117 million at $117,000 a ton, which is pretty close to what he has always said it could cost. And that is with a steel hull. Add in say a quarter the ordinance since we aren’t going to have aegis and a spy-1 radar, or heavy VLS, and your’e at around 300 million. That's around what an unarmed LCS costs but ready to sink subs and escort merchants.
Now the Navy wants a frigate that is probably going to be around 3000 tons which at the same per ton price ends up at 351 million—which amazingly is pretty close to what the base LCS costs and in the ballpark of the LCS tonnage when loaded. So maybe the large size of the LCS for it’s light mission is part of the problem like many critics have said. Add in say 1/2 the cost of the Burke’s ordinance since it isn’t projected to have spy-1/aegis but will have helo support and VLS and you are at around 600-750 million. I may be wrong but isn’t that about where the Navy has put what they want to spend on the new frigate? And that is also what you would charge for a truly up-gunned LCS which critics say is what the Navy wants to buy.
Does this sound about right?
I might quibble with some numbers. The low end ASW vessel, for instance, has some pretty cheap sensors and ordnance. Quibbles aside, your numbers and reasoning is probably good enough for general discussion purposes.
DeleteI should summarize and present the same data for the LCS so we can see how well, or not, the costs scale in a linear fashion. I'll put that on my to-do list. Or, I've given you the reference, you can pull the data out and present it, if you get the urge!
Complicating the interpretation of the Burke cost figures is that it covers two different ships, the Flight IIA and the new Flight III requiring 50% new drawings.
ReplyDeleteTo date only two Flight IIIs have been ordered, one from HII in June and one from GD BIW in September, both as ECP amendments to the 2013 to 2017 MYP contract. The Navy has refused to disclose costs of the contract modifications to the MYP contract for the two Flight III ships.
Originally Navy planned one of its three 2016 buy as a Flight III, but due to slippage not sure it was or both the two 2017 buy are Flight III.
May be but just guessing that the Change Orders at $114M per ship includes estimate by Navy to fund the ECP for Flight IIIs.
Fifteen Burkes ordered since build restart, one in FY2010, two in FY2011, one in FY2012, three in FY2013, one in FY2014, two in FY2015, three in FY2016, and two in FY2017.
Yes, Navy cost data is never straightforward clear. The MYP, for example, further complicates cost discussions and is probably illegal. By law, MYP is only for proven, existing designs which the Flt III is certainly not. The Navy's various budgetary maneuvers and accounting gimmicks make it very difficult to find useful data.
DeleteNo one knows what the change order budget line is for, specifically. Interestingly, the same number appears in the FY2016 budget for two ships. One would think that if the change order was in anticipation of the new Flt III, it would be a one time occurrence rather than a two year, across four ships, occurrence. That level of change order funding also makes it clear that the Flt III is not a repeat of an existing, proven design and therefore does not meet the statutory requirements for MYP funding. Again, the Navy is playing games that are likely illegal.
FYI, the govt contract announcement for the first Flt III specifies that it will be paid for from FY17 budget funds. That makes the FY16 change order budget number a bit of a mystery.
"Navy cost data is never straightforward clear."
DeleteOne recent example as reported by GAO-15-575 in June was the troubled new Advanced Arresting Gear (AAG) which overspent by $572M, a 300% increase above its initial estimate from 2005 . You would have thought as Ford is the first CVN to have the AAG installed the Ford R&D budget would be funding it, but no, it is funded by NAVAIR R&D.
Another classic is the LCS seaframe R&D spend of $4B for the first two ships, the Flight III R&D budget by comparison is only $4.1B, so assuming Navy allocated all the LCS overspends to R&D, they also pulled a similar shell game by funding the new generation RR waterjets for the Freedom class from Navy general R&D funds claiming the new waterjets could be used on other future classes of navy ships, LOL.
The issue of adding length to a ship's hull was mentioned. My naval architecture text book points out that this is the most economical way to increase the capacity of a ship. There is historical precedent for this: during WW2, going from Sumner to Gearing class destroyers, the major change was adding 14 feet between the forward and aft propulsion plants, allowed 168 tons of additional fuel, increasing range by 30%. This was very cost effective way to increase range. The Essex class carriers, starting with the Ticonderoga are often considered a sub-class which had a similar modification. Of course you could go to the extreme and the costs become great: both money and ship's handling.
ReplyDeleteMM-13B