With the yard, in effect, claiming sabotage of a defense program, seems DoJ and DoD would get involved. May as well get the CIA involved too, to see if the sabotage is the work of foreign actors. Yard management better not be trying to cover it’s own sloth and corruption, because what they are claiming would be taken very seriously by every intelligence organization in the US.
In the case of a US Navy shipyard way back when I was there it would have to be a grand conspiracy to fake them. First, the QA experts doing the X-rays have to get ship’s crew permission to bring in the x-ray equipment and get the area safely partitioned off.
Then ship’s security has to allow the techs with the equipment on board.
Usually announcements are made to keep clear of the area being x-rayed.
Then paper worked is signed by the QA and ship’s crew of the completion.
Back then it was film that had to be processed and checked. So if you were faking the actual work you had to go through a lot of effort to make it look like you were doing it.
There was a ~master list of all work on each sub-system that was cross referenced and checked for status, completion and final QA and reported by the shipyard in a weekly (sometimes daily) meeting with the ship’s Engineer, Captain and lead shipyard supervisors. Critical systems were all reported on and spot checked.
Hundreds of issues and requests for re-checks came out of these meetings. Many tests were redone due to questionable/missing paperwork, not all results clearly within range, procedures decided not to be good enough, missing or questionable signatures of ship’s crew, etc.
I don’t weld but one possibility is if the “bad welding” was mostly limited to critical structures.
I do on occasions review scientific papers for publication. An error in a statistical analysis that doesn’t change the statistical significance of a finding is believably accidental. One becomes more suspicious if the “error” makes a statistically insignificant result appear significant.