As I said, I don't know if it is an option. The question in my little universe is the reverse: How do you validate welding?? If it is possible to verify the result somehow you may be able to validate the process, but what do you do when can't validate the result?? Lets say if you build a ship or a house and the welding is being done as a manual process how do you validate that? So, what is my point. If you can validate it, then I believe you must somehow be able to produce objective evidence that the welding is as specified. If you can produce that evidence, and there is a good reason for the decision, then you can use the same kind of evidence to verify your welding.
When you choose - in the medical device industry - whether you need to validate a process, there is a number of considerations:
Is it possible to just verify the process?
Is it economically acceptable to just verify the process?
Does the output of the process pose a danger to the end user?
Is there a salespitch to validation vs. verification?
So, if it is possible to verify the output, and there is an economic or safety reason not to validate then you can choose just to verify.
I don't know if this makes sense, but if it doesn't please respond, and I will try to do better