Many times it is overdone, but you have to look back to the origin of how those specifications generated.
One reason is closing the barn-door after the horses have escaped. Somewhere along the line a nut wasn't tightened to specification, there were only 1.5 thread exposed due to the shortness of the bolt, etc.
I agree with that.
What I have also found is what I think amounts a bit of laziness or sloppiness by those who generate the specifications.
A couple of examples. You know that our area is power electronics generally with quite a heavy emphasis on VSDs. We'll get specs that have been clearly written around fixed speed motor starters.
Another real gem was a huge pile of specifications dumped on me after the contract was placed. The guy staggered into my office with a box of papers. It was a box that we get A4 printer paper in, five reams at a time so 2,500 sheets. And it had great wodges of other bumf taped to the outside of it. So we're looking at probably 3,000 plus pages. And specs that referenced others that weren't included. Made "War and Peace" look like a short tale...
Among this lot was a specification on how to write manuals. There was a whole page on the correct use of the comma. I kid you not.
Second is quality control. Say what you do and do what you say and document it that you've done.
Yes. Part of the manufacturing process. Everything documented from detailed mechanical and electrical drawings, factory acceptance tests, site acceptance tests, certificates of conformity for all component parts, certificates of origin.....etc. And all listed in the document dossier.
We had an 800HP MV drive failed due to an unchecked/faulty capacitor connection that was hidden and not visible without disassembling some panels. The tech' had a hard time finding it, even thouhg he was demonstrably highly skilled. The drive itself is world class assembly from a/the leading manufacturer. The drive was down for days. The short of it is that there were no recuperable damages, production loss, whereas if a specification had detailed checklist for connections and such records were asked to be furnished, the likeliness of this type failure would have been reduced.
Again, I agree. In the works we have an inspection procedure that checks all connections and they get marked with a felt tipped pen when they have been. Not foolproof of course - you could have someone on a bad day just marking them off and not checking.
I think ISO9000 series went a long way so that you CAN audit the manufacturing QC process and only write into your spec that is missing from their process, but you think is important. Often times during the Factory Audit the Manufacturer will revise their ISO process to accommodate your interest.
We are ISO9001 certified. The main thing is, as you say, document what you do and demonstrate you do.
Tomorrow we have another quality audit by a different body that we need to keep us on the list for being kept informed of contracts being let in the utility sectors.