A New Tool to Analyze the Shipbuilding Plan
15:02 Share

This is CNA Talks, the flagship podcast of CNA, a Washington, D.C.-based research and analysis organization. Ships in the U.S. Navy take years to construct, and maintaining the fleet requires planning decades into the future. In this episode, we examine a CNA-funded dashboard that lets U.S. Navy commanders examine the Navy shipbuilding plan and see how hypothetical changes would affect the fleet.

Today, we're going to be discussing a CNA tool that's been developed to assist with the Navy's shipbuilding plan. And I'm excited to welcome a first-timer to the show to discuss this with me. Rachel Hartley is a research analyst with CNA's Force Optimization and Readiness Generation Program. She specializes in addressing recruiting crises in the military and performance-to-plan analysis, which has been used in tools like MyNavyHR. Rachel, thanks so much for coming on the show today.

Hi, John. Thank you. So let's get the context out of the way first. What is the Navy Shipbuilding Plan? Just tell us a little bit about that. Yeah. So every year the Navy publishes this paper. The official title is the Annual Long Range Plan for Construction of Naval Vessels. But that's a bit too much for most people. It's usually just referred to as the Shipbuilding Plan.

And it gives sort of the Navy's schedule for ship construction and decommission over the next 30 years. It gives a little bit of explanation for those decisions and gives a few charts and tables that kind of summarize things like ship count and cost. Got it. And so in this case, what challenges were you hoping to address with this project in regards to the shipbuilding plan?

So the shipbuilding plan, most of the details in there are more text-based. So there's lists of which ships will be decommissioned each year, but not really much information on how

how that directly affects things like ship count or costs, right? So it doesn't say you start with this many ships, we're losing these, and you end with that many. You kind of have to pull all those pieces together yourself if you're interested in that. The few graphs that they have in there are very...

and kind of almost unreadable if you're interested in details about, you know, how much is it actually costing per year to keep all of the submarines going in the Navy. It's also, you know, difficult to tie all of that data to things that are not specifically mentioned in the shipbuilding plan, like manpower data. Like you mentioned in your introduction, I've worked a lot on manning in the Navy and

So the first time I saw the shipbuilding plan, I was kind of interested in, okay, well, you have this particular collection of ships. What does that mean for how many people the Navy needs to have at sea every year? And so trying to tie those pieces of information together takes a good while and a couple spreadsheets to calculate in the current status of the shipbuilding plan.

Got it. So it sounds like the plan is kind of opaque to anyone who hasn't spent time building it. And, you know, while I'm sure there are people in the Navy who this is everything they do, you also probably have quite a few people who, you know, that's not what their specialty is. And they just want to be able to quickly understand, like, what the impact of this plan is, what the goals are and how different, you know, decommissions or construction, what that'll mean for the forest structure. And your tool helps make those kind of things easier.

Exactly. Yeah. It kind of takes the whole thing and tries to summarize it in a nice and interactable way. And so this tool is called the Shiny Code Package. Is that correct? Yep. It's a, like you said, a package that was originally developed for the R programming language, but now also works with Python. Got it. And can you tell me a little bit about like, you know, how it works and the development process for this? Yeah. So it's basically like a...

giant graphical user interface or GUI package where you can choose absolutely everything about how a dashboard is set up. You can choose graphs or tables or almost anything you can imagine. There's probably a package to help you do it. Network graphs,

Kind of the big advantage it has over other dashboard programs like Tableau or Power BI is that you have all the power of the coding language behind it. So you can actually have your user enter information, you can pass it to your code, do some analysis, and then display it back to your user and almost instantly.

Yeah, that sounds very useful, especially for people who are higher up in the Navy or in Congress, perhaps, who don't have a lot of time to go through all of the details of this. This just gives them a very straightforward, visually appealing way to engage with all of this information. Exactly. And so not only does it help you visualize what's in the plan itself, but it also will let you make changes and see how that affects the

you know, different things like cost, manpower, the number of each type of ship you have for each particular year. So the app was kind of started as a way to, you know, revisualize the shipbuilding plan and in turn kind of ended up being a way to sort of analyze it as well. It almost seems crazy to me that nothing like this existed before. It just seems like such an intuitive thing that like would be so useful to so many people in such a massive bureaucracy like the U.S. Navy. Yeah.

You know, it's really hard to plan 30 years into the future. So most people take the first year or two as pretty good to plan off of. But what's going to happen in 20 years, I think most people know that's going to change and probably change a lot. So they tend to pay more attention to the first year or two and maybe not so much the other 28 years.

Mm hmm. That actually leads me nicely into my next question, because you're talking about this long view of idea of shipbuilding. And on a recent episode of CNA Talks, I talked to a few of our analysts about the AUKUS agreement and how as part of that agreement, the US is seeking to ramp up its shipbuilding. Can you just talk a little bit about like how, you know, that kind of analysis over time can maybe maybe is different than the short term stuff we've been talking about? Yeah, I think one thing that

You might not think about too much if you're not involved in the shipbuilding process is just how long it takes to make these ships.

Especially if you include all of the design phase. I think most of the ships in the model, just the actual building time is somewhere between five and ten years. So if you take five years to plan a ship, ten years to build it, that's already thinking 15 years out into the future. So in that context, 30 years doesn't seem quite as long. But most people are just not used to thinking about that part of the process.

It of course does make sense that it takes that long. These are massive ships and like, you know, I've heard multiple times aircraft carriers be referred to as floating cities. But if you're not involved in this process and you're just used to these things being around, it isn't intuitive that it takes that long to do that, to build something like that. Are there any other places in the Navy or elsewhere in the U.S. government that a tool like this could be applied?

Oh, absolutely. I think pretty much anywhere you have data that you want to analyze or summarize, a Shiny app would probably be useful. We have quite a few others that we've put together in the data science department for the Navy. I've worked on another one.

That visualizes the supply chain for Super Hornet parts for the CNAS in San Diego. And there's another big one called ADM. I forget what that acronym stands for, but it helps Opnev and 832 work on their POM planning process. Both of those are hosted on Jupiter in Edvina.

Just to help our audience understand how this tool works a little bit, could you just talk us through a hypothetical example of how one might use it to answer some questions about the shipbuilding plan? Yeah, absolutely. So when you open the app, it loads in the configuration for the default shipbuilding plan. And then you can go in and change when ships are being commissioned, when they're being decommissioned. So say you wanted to...

change when a particular set of ships gets decommissioned and move up that timeline. So take the littoral combat ships. So the shipbuilding plan has the LCSs start retiring in about FY32. Say we move that up to FY28, the graph on top will update and show the total inventory of the ships start going down a little bit sooner.

And if you compare the two lines, you can see it actually takes quite a bit longer to recover back up to the value of ships you had before you started those decommissions than in the original plan, because now you kind of have a gap of ships that they were relying on that doesn't get filled in. So theoretically, you could go and start playing around with

moving up another production of ships to try to fill that gap or trying to follow the consequences outside of the app, what would that do? Would you not be able to complete certain missions because you don't have the ship power for it? Right. So I can definitely see the use case of that because now you're able to answer questions and follow through the consequences of

of a decision, you know, what maybe in the immediate term feels like, oh, these are, you know, dated ships. We're working on replacing them anyway. Why not decommission them sooner? But, you know, you pretty quickly see the consequences of what that would look like. And those kind of visualizations just seem like they would be invaluable to making, like, informed decisions here.

Absolutely. I'm not really much in the shipbuilding world myself, but I've started thinking about a lot of those types of questions just working on the app here. Yeah. So just overall, Rachel, can you just give us the big picture of what listeners should take away about this tool and what you want them to understand? Yeah. I think it's a great tool that can be used in almost any situation that involves data analysis.

Of course, not every project needs a dashboard, but because you have the power of R behind it, you can do things like upload an Excel sheet, have the computer do all the analysis for you and spit back out the answers rather than having to sort of fiddle with it yourself.

If you do need a dashboard, it's super customizable. You're not sort of limited to the pre-built things that Qlik or Tableau has. And one of the great things is it works on Jupyter, which is the Navy's current cloud computing environment. I would have gone with Neptune, but fair enough. Oh, yeah. It has...

tons and tons of data up there that you can pull in with R and then display with R Shiny, sort of all a one-stop shop for all of your data needs. Okay, Rachel. So just to wrap things up, this project is part of CNA's larger Innovation Incubator Initiative. And I just wanted to ask you a little bit about what that kind of internal funding that comes from our company as opposed to clients or sponsors means for your ability to create new solutions to our sponsors' problems.

Yeah, it's a really interesting position to be in. You know, the Navy is always so busy with all of their missions and their money is not unlimited. So with our self-funded projects, it's a really nice way for us to sort of pitch some ideas that the Navy, you know, might not have the resources to fund or might not really have the time to explore on their own. And we can sort of

pull a bunch of different strings that we see between different departments and make some connections that, you know, wouldn't be impossible for the Navy to do on their own, but just might not be able to be a priority. Yeah, I mean, it's one of those things that I come back to a lot speaking to the team here that works with the Navy is our relationship and our long standing connection to the Navy does give us like a unique insight that I don't think a lot of other FFRDCs have.

So it's great that we can take the initiative to address some of these things with CNA funding. Okay, well, Rachel, we're just about out of time for today, but I want to thank you so much for coming on the show and telling us about this exciting product of yours. Yeah, thank you so much for having me. You're interested in taking a look at some of CNA's self-funded projects. We'll have a link to our Innovation Incubator webpage in the show notes of this episode, so you can go check that out. But thank you all so much for listening, and we'll see you next time.

on CNA Talks. The views expressed are those of the commentators and do not necessarily reflect those of CNA or any of its sponsors. CNA Talks is produced, edited, and mixed by John Stimson. Our theme music is by Edward Granga. If you enjoy our show, we'd love it if you could give us a five-star review on Apple Podcasts and tell your friends about us. Thanks again for listening, and we'll see you in two weeks. ♪