Home » Budget Industry » Geurts: Digitized Design Is Key to Speeding Up Shipbuilding, Repairs


Geurts: Digitized Design Is Key to Speeding Up Shipbuilding, Repairs

Siemens PLM Illustration

WASHINGTON, D.C. — Digitization is increasingly serving as an important tool to drive down shipbuilding and repair costs, which will be vital to the Navy’s ability to afford its future larger fleet, the Navy’s top systems-buyer said on Tuesday.

With about 100 major repairs performed on vessels each year, the Navy has to find ways to save more money when it comes to maintaining the fleet. Otherwise, the Navy will not be able to afford the planned fleet expansion to a projected 355 ships, James “Hondo” Geurts, assistant secretary of the Navy for research, development and acquisition, said while speaking at the Defense One Tech Summit today.

“The talent is there, the need is there, the challenges are there. Quite frankly, if we don’t do it, we aren’t going to succeed as a nation,” Geurts said.

James F. Geurts

The National Defense Strategy does a good job of setting the Navy’s priorities, and the Navy and Marine Corps leadership have detailed their vision of how the service should satisfy these priorities, Geurts said.

The challenge, Geurts said, is quickly delivering lethal capability but in a way that doesn’t involve runaway program costs. Using digitized design is part of the solution, he said, because the technology helps manufacturers speed up production.

In some cases, workforce training is improved by providing a clear digital picture of what is being built, Geurts said.

Digital modeling can also cut time out of the design process by allowing engineers to test capabilities in a matter of days rather than needing weeks or months to work out solutions through a series of trials, as occurs today. The Navy is trying digital design methods with the MQ-25 unmanned aerial tanker program, hoping the work will speed up the time-table for fielding the capability,  Navy officials have previously stated.

Leveraging digitization is already happening in Navy shipbuilding, Geurts said. Digital ship plans can cut down on the number of hours required to build ships. Digital systems can be upgraded quicker by sending out software upgrades over the air instead of waiting for ships to arrive in port for upgrades.

“For the first time ever we have full-up, from the start, digital ship designs,” Geurts said. “Two of the world’s most complex digital designs right now are for nuclear aircraft carriers and nuclear submarines.”

The Huntington Ingalls Industries’ Newport News Shipbuilding yard, builder of the Navy’s nuclear aircraft carriers, used a paperless design for the third Ford-class carrier, the future Enterprise (CVN-80). The company has previously said such digital designs could result in saving more than 15 percent of the overall shipbuilding costs.

Digitization helps with the design work, but Geurts said digital plans create a wealth of possibilities to save time and money. Digital plans can change how ships are built, how they’re inspected, how they’re maintained, and how they’re upgraded.

The Navy is currently running a pilot program using data collected from the surface warfare community to make ship maintenance availabilities more efficient. Instead of ship’s supply officers running down a check list of parts to be replaced when in port, the Navy is testing a process suggesting replacement parts be considered based on historical data. For instance, if a blade needs replacing, the data-based system suggests to the supply officer bolts commonly used with the blade and any other components that tend to need repair or servicing at the same time. The idea is maximize time spent doing repairs and only spending money on parts that are needed.

Referring to an ideal goal for creating quick computer system upgrades, Geurts said, “the challenge is 24-hours from compile to combat.” In what Geurts described as ongoing tests, the Navy has already shown it is possible to compile software upgrades, test them, get them cyber certified, broadcast them over the air and to ships at sea, have them uploaded in a combat system, and begin operate the upgrade within 24-hours.

“I firmly believe we can innovate at scale. I don’t think that you have to a ‘special’ in front of your name or a ‘rapid’ in front of your name or something in order to innovate at scale,” Geurts said.

  • NavySubNuke

    Overall I think this is a great thing for the Navy and the Nation.
    The biggest issue to be resolved though is the cyber security aspects. The cyber aspect involves not only protecting the design from compromise and reverse engineering/copying but also extends to keeping control over the configuration of the design itself so we don’t waste time and money based on unauthorized changes to the plans after the ship is build. What I mean by that is how do we ensure that 10 years from now and 20 years and 30 years – the digital designs we are using to plan upgrades and overhauls haven’t been tampered with and are still accurate?
    Also, how do we make sure 30 years from now we can even open the plans? How many data files from 1988 can still be opened right now without major formatting errors or other problems? And I don’t mean word documents – I mean something technical like the design of a 100,000+ ton aircraft carrier. Making sure the software stays up to date and the design doesn’t mutate as software version change is going to be a real challenge as well.

    • PolicyWonk

      The key to retaining data currency and accuracy through the years, to avoid the compatibility problems you mention, is to use an underlying storage system (i.e. database/data management platform) that is maintained via a upgrade program, thereby always staying current. Software upgrades, etc., for the supporting application, in this case computer-aided-design (CAD), also must occur on a regular basis, as do the operating systems, etc.

      What I have seen, typically, are large organizations as part of the MIC that think they’re saving money by not bothering to do their upgrades on time, or by forgoing license fees, support costs, and maintenance, etc., because “it still works even when you don’t”.

      But systems almost always go south, or have some technical glitch (Murphy’s law applies to computing systems, just like anything else). And when they do, it can be crazy expensive (and time consuming) to get them recovered, and up and running, if/when their upgrades/support/licensing and systems haven’t been maintained.

      This happened to one of my DoD clients: When I arrived I was immediately marched (with 4 security guys marching alongside me on both sides) through the facility, with literally thousands of people (engineers, technicians, machinists, etc.) standing still, watching me get escorted to the systems area. This massive facility was as a complete standstill (we walked for about 1/2 hour indoors, to get to this system rooms), over a $20k machine and software system that hadn’t been upgraded in more than a decade, that failed at a very bad time. The money that was “saved” over the previous 10 years evaporated in about 5 seconds, and this system was offline for almost a day (the operational cost of all those thousands of highly compensated people, the machinery, facility, etc., must’ve been staggering).

      You think the vendor gave that company a discount for upgrades to their licensing, etc, after that?

      Note that in the financial world, the trading systems in place are the same COBOL programs that were written back in the 1960’s, and are still running and maintained to this day: so the problems you mention are entirely avoidable. Financial systems that run the economy cannot fail, as billions of dollars can be lost in the blink of an eye. The DoD/MIC can also learn how to do this – but in a lot of ways technology simply isn’t their strong point (ironic, considering how advanced some of our weapons, etc., are).

      You can’t afford to have a system like this being overseen by people who aren’t responsible technologists, who are pretending to “save money”. And this is unfortunately typical for the feds, and MIC/DoD installations I’ve been involved with.

      • Zorcon, Fidei Defensor

        I have made 50% of my career keeping old DoD systems running. Don’t bad mouth my food chain 🙂

        • PolicyWonk

          Heh –

          Been there – done that! It’s sometimes been part of mine. Clearly, you know what I’m a-talkin’ about 😛

          • Zorcon, Fidei Defensor

            All to well. Years ago, I got a call about a ATE test system. This goofy old design booted a old DOS PC system and then loaded HP Basic (remember that one AKA Rock Mountain Basic) onto a ancient 68000 based co-processor card in a ISA slot. Pretty much a attempt to make the old HP9836 computer a PC.

            Said system had a small, obsolete hard drive that had a MSDOS partition and a obsolete and pleasantly absurd HP proprietary format partition. Said system had a clone at a manufacturer. Thus only 2 drives existed on planet Earth and one drive decided it was time to retire causing panic and abject chaos. Imagine that, after 30 years, dying.

            In a nutshell, had to find reconditioned drives (did you know that a old tiny DOS HDD is many times more than a new 1 tb drive?) and clone it. Of course, cloning involves more than Norton or some such program which is worthless for anything other than NTFS and such.

            I bet, to this day, that system is still being used.

          • PolicyWonk

            Yep – last year I had to do some work on a system that was old enough to go to (and probably graduate) college. This is forever in technology terms.

            Remember the UAV that Iran hijacked and landed on their territory? They rushed it into service when it was completely sans any security in the controlling SW. Whoever forced the approval though should’ve been drawn and quartered.

            You might recall the infamous Bradly Manning debacle. The SW he was using to access 750k files (given to wikileaks) allowed him to see data that was far out of his area of operations. No one detected that he’d downloaded GB’s of data to removable storage (i.e. a CD). In the financial world, such an action would’ve set off alarms from Boston to Baghdad, because the SW that tracks such things has been available for decades.

            The USA does a lot of stupid things to itself, while pretending they’re saving money. Manning went to prison (and deservedly so). But no one was court-martialed for dereliction of duty for failing to secure the data. In short – that breech never should’ve occurred – and/or should’ve been discovered the second he started copying data to removable storage.

            The MP’s could’ve been there to take him away before he finished his shift.

          • Zorcon, Fidei Defensor

            Don’t leave out Reality Winner. At least she got caught.

          • PolicyWonk

            What she did was simply dumb – and she got caught because of a printer Id. Compared to Manning, she was small potatoes.

            Manning was less of a big deal than many would like to admit. Mostly, the stuff he gave to wikileaks merely confirmed what those of us who follow military and foreign affairs already knew. That idiot didn’t know he was giving away info that was already published globally.

          • Zorcon, Fidei Defensor

            That he got commuted and is off making a idiot of “himself” is what really flames me. Think I would have skated for 1/5 that transgression?

      • Curtis Conway

        Back in the day, we were taught to be as efficient as possible in our code writing, proving and implementation. Today most all of that is taken care of in Compilers, and in some cases you have software writing software. There is so much vapor-ware in the Operating System, much of which is not understood, it is very easy for malicious software to hide out and be launched at another date. Certified systems are supposed to TRANSCEND these anomalies. Guess not, all to save a few dollars in a $Billion/Trillion world. Go figure.

        • PolicyWonk

          In many respects, the DoD should not be using COTS operating systems, that may have “Trojan Horses” built in, specifically due to security concerns. Versions of operating systems for secure environments usually run bare-bones, without all the additional c r a p-ware. You want the minimal amount needed to get the job done, and nothing more.

    • NEC338x

      Making sure software is up to date and harmonized to database files, tracking every post-design change and ensuring its entry and distribution across databases, and doing so consistently when a ship design will outlast the average human employment period is hard, adds to costs, and risks non-recovery of data from a single point failure.

      • NavySubNuke

        Agreed. Even without the cyber security concerns of ensuring we do not allow an adversary to compromise the design – it is a non-trivial problem for sure!!

        • old guy

          Just keep it OFF LINE. Problem solved.

          • NavySubNuke

            Not in the slightest. Let’s not forget BUCKSHOT YANKEE – a now unclassified example of the SIPRNET being penetrated.
            Also, according to published reports, STUXNET was able to get into and destroy the Iranian centrifuges despite them being offline.
            Air gapped networks may help the problem but they in no way solve it.

    • Chesapeakeguy

      Adhering to ISO standards should ensure that the ‘latest, greatest’ is being used and implemented. SHOULD….

  • Sons of Liberty

    Yes by all means thai isnthe reason we have cost containment issues. Wasn’t the Ford a vetted “digitized design”?

    One has to wonder how if these guys are chinese sleeper agents ir just a 5th column. Looking to digitize all efforts so then can be easily download by China.

    • USNVO

      There seems to be a little confusion in the article. If you follow the link, the previous article did a much better job explaining. The FORD was a completely digitized design, but the workers built the ship with paper plans. The ENTERPRISE is supposed to be constructed without the workers using any paper plans at all. So paperless construction, not paperless design.

  • Marc

    I thought we started doing this on all new ships designs 20 or 30 years ago? Why is this a great new idea?

  • Zorcon, Fidei Defensor

    Wow, welcome to the 20th century. Boeing started this decades back with the 777 and SolidWorks is the tool of almost all small business.

    So the Navy is 30 years behind the curve? Look at the upside though. The Chinese cannot steal paper drawings by hacking the servers.

    • PolicyWonk

      Heh –

      Think about the amount of storage a full-blown carrier design (with versioning!) would take. To create a CAD with that kind of scalability and decent performance would be an impressive feat.

      • old guy

        Less than one hour’s requirement for Facebook!

        • PolicyWonk

          Different purpose, application, requirements, and data foundation.

    • old guy

      40 not 30 years ago.

      • Zorcon, Fidei Defensor

        Well, I was going off the first flight of the 777 which is when I left. That would have been 1995. They used CATIA on that airframe. So, that is about 30 years?

    • old guy

      BOEING was one of the companies that received the CASDAC program.

  • NEC338x

    There is a place for digital design, but as I make a point to impress upon my undergrad engineering students, you can easily over-design things with AutoCAD and SolidWorks. That’s why I make it a habit of running a drawing package past a machinist or two before I ever send it out for bid. The perfect package still must end in bending metal/building circuits/writing code. That’s where the costs will start piling up.

    As to modeling, again software packages are great as long as you realize you can model right through unknown-unknowns and never catch them until Murphy points the failures out. To quote H.T Odom from Modeling for All Scales, “It is nonsense to try and model all the detail of the real world. To do so would be to lose understanding in the confusion. It takes more information to model something than to be that something. To find out what the real world does, measure the real world. To understand and predict the main features of the real world, develop models on a scale appropriate to human understanding, with frequent checks with observed data…” Synergistic effects will only be discovered when observations start to deviate from the predictions.

    There are appropriate roles for modeling and appropriate roles for testing. Don’t throw the baby out with the bathwater because someone is flipping through a colorized PPT, and its the most recent shiny object. The default response to a question by a program manager should not be “lets do some more modeling”, which is a risk.

    • Zorcon, Fidei Defensor

      My mechanical integration guy IS a machinist as well as a ME with decades of experience.

      Designing things is one thing. Deigning them to build is another.

    • PolicyWonk

      An excellent point. Having spent most of my career in engineering environments, you can draw all you want, but you still can’t machine around corners ;-P

      Great engineers have a lot of common sense – the others are a waste of time.

      • Lazarus

        So you are an Engineering and not a policy wonk.

  • Michael Hoskins, Privileged

    I used this sort of technology in industrial process design and construction during the last 20 or so years of my working life.

    Additionally, about 15 years ago I was able to implement commercial, off the self technology to scan existing significant National Asset Working Buildings in DC to create digital drawings to sub millimeter accuracy for remodel and repair.

    For less than $50k we bought an R2D2 looking device, about 30” high, that scanned 75 year old attics, sub-basements, architecturally significant lobbies and much more. These systems are now in very mature use throughout the major projects construction industry. The accuracy has allowed some pretty amazing construction techniques.

    Several such systems could by dispatched from ship to ship as needed for pre overhaul planning and upgrade design. Flowing this back into original digital designs could create a data base of “as built” files on a ship by ship basis. New designs could more easily evolve from that point.

    And, it is proven COTS technology and skill sets.

  • Lazarus

    Digital design has been around for a while and was used in the LPD 17 design. Costs still escalated. The real question is how do the shipbuilding components of Defense contractors who are in monopsony relationships with the govt find ways to reduce cost despite Congressional budget instability. Wonder why the shipyard force is inexperienced? It costs big $$$ to keep those shipbuilding lines stood up while Congress figures out what it wants to do. US military shipbuilders do not have non military projects to which to allocate these workers when Defense funding is lean. Few options for achieving cost reductions in this environment.

  • old guy

    40 years ago my guys in SEA03 created CASDAC [Computer Aided Ship Design and Consruction} The world picked it up and it became the well known CAD’CAM [Computer Aided Design-Computer Aided Manufacturing}, 3 guesses who did not require it in new shipbuilding contracts. Being 91, I can’t recall his nsme, but I know that I have it somewhere. I will look.

  • Chesapeakeguy

    So long as it works…

  • Secundius

    The “Gerald Ford” was a Virtually Computer Design” from the Keel Up! And see where that took us, when Virtually Reality meet Head-to-Head with the Reality of “Murphy” (i.e. the Real World)…