Hello Chris.
First, tests are done on the whole a/c, not the wing only or at least that is the usual case. It would be insufficient to test only one element of the whole structure.
From what I have seen you modify the existing test article (i.e add the straps) and redo all the critical loads. Also, it is possible to wait for the next available a/c that will have modified wing (highly unlikely, it can work only if the issue is minor, which is not the case at the moment).
What you mention as computer modeling can help, of course to avoid modifications, only in one case: when the appropriate certification authority (EASA in this case, FAA with them on board, I would expect) accepts that argument.
But that already goes into willingness on the OEM side to step in and fight for its solution. It seems to me that it was easier to "strap" the weakest portion of the structure and continue your work. Not a big deal, at least not for those on the structures/engineering side.
However, for general public, it might look as an embarrassment.
I do not know of any typical percentage rate or any correlation for the repeated tests. Maybe someone else can give you better answer.
Cheers