The use of digital models and generative-AI tools could become standard in Army contracting if the service can clear hurdles related to workforce training and trust, the deputy assistant Army secretary for data, engineering, and software says.
It’s part of the service’s effort to make digital engineering ubiquitous in all aspects of acquisition. In May, leaders released a directive outlining the plan and tapping six weapons programs to pioneer the broad use of digital engineering tools. The two largest programs—the Future Long-Range Assault Aircraft and the XM-30 Mechanized Infantry Combat Vehicle, were both “born digital,” which is “fantastic,” Jennifer Swanson told reporters at a Pentagon briefing on Tuesday.
The leaders of both programs have worked for several years to train their teams on digital tools and their use in contracting language, Swanson said.
“We’re using these programs to help us take their lessons learned, what is working, what is not working, and be able to adapt all of that for more of an enterprise view across all of our programs to really help everybody else,” she said.
“The other communities that we want to make sure are operating in these digital environments— like the requirements community, that test community, the sustainment community—we’re not there yet,” Swanson said. “Those Pathfinders will still benefit a lot from being able to expand this concept in these environments across to those other communities so we can really build out those digital threads.”
About 20 Army programs currently use digital engineering, Swanson said.
By integrating digital twins and models into contracting practices more could help the Army better communicate what it needs to industry.
“Digital twins are fantastic because they allow us to replicate capabilities that are in the field, in the lab digitally or in a computer digitally, and really be able to replicate problems, identify root causes and problems. It just allows us a lot of speed and agility and insight that we would otherwise lack,” Swanson said.
The Army aims to have more contracts and programs that require digital engineering from the start.
“We also want to use digital models to go out with [requests for proposals] to determine conformance, or compliance with our own standards,” and have vendors respond with their solutions so the Army can identify whether it actually fits with what the service wants, Swanson said. “Because that is something that is honestly impossible to do manually. And so a lot of times we’ll require standards and RFPs and…we do the best we can, but we don’t really know before we award if it’s really, really going to be totally compliant. And so I think digital models will help us to do that much more.”
But there’s still a learning curve with digital tools for contract workers, one that could cause program delays. In its 2024 annual weapons systems assessment, the Government Accountability Office found workers’ inexperience with digital engineering slowed contracting efforts for FLRAA.
The program is working through two challenges with developing digital twins: “ensuring the use of open, adaptable, and secure digital engineering tools; and providing secure access to the data/models to stakeholders that need it,” the report said.
For the XM-30 program, officials said it “took longer to release the request for proposals due to a lack of experience with digital engineering while directing contractors to use specific software design approaches” the report stated. Also, there wasn’t a precedent for a “digital open architecture project, which delayed the source selection and evaluation board process.”
Swanson said those slowdowns were growing pains of being among the Army’s first programs to try digital engineering. But there’s an added problem: everyone uses different tools that don’t necessarily work well together.
“It’s a big problem,” Swanson said. “You have the [program manager], who has established their own digital engineering environment…vendors have digital engineering environments as well. The Army is not going to direct everybody to use a certain tool—that doesn’t make sense for us.”
But to prevent slowdowns in contracting, the Army is planning to train the acquisition workforce in digital engineering with online and hands-on classes.
“If you’ve ever looked at a model produced by digital engineering tools, it is not like looking at a PowerPoint chart. It requires you to think differently and really adapt the way that you’re used to doing things. And so of course, with that comes, it takes a little bit of time,” she said. “But I think our workforce is ready. And we’ve certainly been peppering them with a lot of digital courses over the last couple of years for software and data. So I don’t think this is any different.”
The Army is also planning to start a generative AI pilot this summer that could help with contracting, Swanson announced at Defense One’s Tech Summit on Tuesday.
“So generative AI. That’s a thing today that’s out there. Not as much for the government, though, because we are worried about our data,” Swanson said. “I will be kicking off a generative AI pilot this summer with ASAALT with a tool that does have an [authority to operate] for [Impact Level] 5,” which is for sensitive, controlled unclassified information, she said. “I’m excited about that because I think it can be a productivity booster.”
While generative AI likely won’t be immediately used in Army contracts, Swanson said there was a lot of potential to boost acquisition workers’ productivity for back office tasks, business systems, contract policy among other areas.
“As a pilot, it’s not just about increasing our productivity, which will be great. But also what are the other things that we can do and then what are the other industry tools that are out there that we might be able to leverage or add on?” she said. “So might one day be able to write a contract? Hope so. But we’ve got to pilot and test to make sure everyone is comfortable.”
