Breaking the Bottleneck: Re-industrializing Defense through Smarter Testing

America doesn’t build ships and submarines on time, at a reasonable cost, or in sufficient quantity. These are critical defense systems, and transforming speed and output - while controlling costs - is critical to national security. Where should industry leading technology be inserted to solve this problem?

One area for high impact is component testing across the supply chain. This seems like a rather niche problem until you realize that broken legacy testing methods are responsible for supply chain bottlenecks at every stage of the defense system manufacturing process, from raw material forging and casting, to component machining, to subsystem development, final system integration, and beyond. As components haphazardly stumble through the supply chain, they become paralyzed at choke points, unable to move onto the next gate in a timely way.?

Of course, critical defense systems must undergo rigorous testing to meet DoD’s high accreditation standards. But the current testing methods simply don’t meet the testing need on operationally relevant timelines. The challenges are legion: manual, slow, and scarce data collection provides inadequate visibility into component health, tangled and confused supplier coordination across the defense base spawns delays, flaws caught far too late cause continual costly rework of components, and a tiny workforce bereft of industry-leading technology? is unable to address the sheer scale of the testing backlogs.?

The result? Legacy methods have created a situation where components languish in forges, foundries, shipyards, and depots for weeks, months, even years—each delay compounding the previous ones, repeatedly blowing past final system delivery dates. Actual fleet counts fall far short of desired and stated fleet counts.?

The good news is that these challenges can be addressed to a very significant extent through existing commercial capabilities for strategic automation and digitization. This includes robots doing difficult, dangerous, and dirty data collection to augment (not replace) the inspection workforce, with data empowering inspectors to make faster and higher quality assessment decisions; creating a clear digital thread of component testing at every gate in the supply chain accessible to decision makers; and building entirely new categories of software-driven workflows to accelerate the testing data aggregation, annotation, review, and approval process.?

These approaches provide striking potential leverage against the supply chain malaise. Often, similar structural reasons create paralysis at different gates in the supply chain, so solving or ameliorating the issue at one gate provides a replicable and faster-to-deployment model for other gates. Moreover, while it is true that testing delays compound, the inverse is also true: the right kind of testing improvements have a positive spiral effect, creating throughput improvements much greater than anticipated, and offering opportunities for drastically reducing or even eliminating costly course corrections in the manufacturing process downstream. And finally, a data-centric approach that preserves the digital testing history in perpetuity means that we get to uncover and operationalize manufacturing learnings as we go, localizing sources of problems, fine tuning testing thresholds at the level of a single component or an entire class of components, improving pattern recognition, and more.?

There is a broader point here. Re-industrialization is not just about simply building more platforms, or adopting new technologies that hitherto were excluded from the defense industrial base. It should be about holistically revitalizing the processes that we use to build, and unblocking the hurdles that currently constrain them.?