Price of Nonconformance (PoNC) - Telecommunications Service Industry

[The Fixer]

We are in the Telecommunications Service Industry. (specifically repair and resale of all the "Baby Bell" systems equipment as well as independant phone companies)
My question is how far do I go in measuring the Price of Non-Conformance ? Right now, our primary inputs are rework due to QC inspection failures and warranty work.
Obviously, you could argue that the receivers wouldn't have had to receive the unit if it worked the first time,then there's the tech's time, the shippers time, etc., etc..
I don't want to get to the point where we are tracking the number of pencils used during the process of handling non-conformances (not that anyone really uses pencils anymore) but I dont want to keep the scope so broad as to render the quantitative analysis ineffective.
Can you throw a few tidbits out here for me to consider ?

The Fixer,
I'm not quite clear regards your situation. It seems that you are part of a group that repairs warrentee and other failed units which are delivered to your receiving section, fixed and shipped out and sold. What isn't clear is what business unit you are part of. Does the above description describe the whole business unit?
Certainly the above activities are all part of someone's cost of quality/price of nonconformance. But whose?
Your group's cost of quality is not the cost of these repairs. It is your primary business. Your cost of quality is the cost of the ones you think you fixed but which are returned or fall off your line through your own failure to meet your own requirements.
Elsewhere there is a discussion on cost of quality. If I'm true to the position I took there I would say; "don't measure your cost of quality. Look at the yields. How many returns did we get that we thought we had fixed? How many recycled through our lines because we didn't do the job right? Are we good at identifying the returns and recycles that are our fault and those which are our customer's fault and do we have a significant number of 'no fault found' where we can't be sure if we've missed something or if the customer has wrongly diagnosed? To try to establish the cost of these issues does nothing to solve them. We need engineers and techs to analyse the process and take out the errors, and hense, take out the cost, whatever it is. Convince management that you are doing the right thing and doing it well. If they want a cost, then say, our yield was 95% (ie; 5% recycles plus returns which were our fault. Get finance to tell you what your cost of sale is, or your operating expenditure for a month, year, etc. Calculate 5% of that and give them that figure. Then they will say; 'That's not right. We're all here anyway, the plant and equipment is paid for anyway, the canteen staff are on duty, etc, etc. But that isn't the right way to do it. You could check this against the standard cost and actual cost for a month to see if things seem to add up, but I wouldn't take too much notice of the results, you could start something that would lead you in circles. 5% of your costs is lost in rework. Whether they could or could not recuperate that money is another question. I know they can't take away 5% of a canteen person. But 5% of cost of sales is a valid figure and it's a lot of money. It probably understates the true cost by miles, because failure creates havoc in the process, but it should be enough to convince people that you need to be supported in solving the problems, especially if they are told where you are coming from. Don't start measuring and totalling, because you are needed out there engineering the process.
Alternatively you can count the pencils and all the other stuff. Do you know that people never use a pencil down to less than 5 inches on average? They lend, lose or throw them away long before they are even half used up. How are you going to factor this information into your cost of quality data? Tough one, eh? What has this to do with equipment failures? Nothing? Everything? Who knows?
Best to go back to the line and eliminate the problem. Don't count it, fix it.
If it is your customer that wants a figure, it's in your invoices. Total for the month, year, etc, plus their internal costs of replacement, administration, etc.
One more thing; How do you know what to charge for out-of-warentee jobs? There's one to think about.
I know this is a bit of a mixed up sort of answer, but you asked us to throw ideas out and it's a typical friday afternoon, after work type of offloading. Best of luck.
and rgds, John C

[Marc]

The main Cost of Quality thread as of this date is: http://Elsmar.com/ubb/Forum5/HTML/000001.html

There is also an 8-D pdf file in the pdf_files directory here you might want to look at.

[The Fixer]

I already replied but it didn't post. If both replies show up all of a sudden, just delete one.
John C.,
Thanks for the insightful reply. Your strategy of examining yields is logical. I think that's where a difference can be made through Internal Audits and Quality Action Teams. I do have to specify some tangible cost reduction evidence to get the bad taste of Quality out of the CEO's mouth. We do a fantastic job of passing Bellcore and KEMA audits but he's not convinced that the Quality system is a viable resource for increasing profit margins. Warranty rate reduction and process/testing improvements based on NTF rates (as you suggest) have not been aggressively approached as part of a realistic CI program. I am in a position now where I have the leverage to drive some of these programs and if even an "outsider" like yourself can see the wisdom in targeting these areas, I am even more convinced that's the way to go. My present thought is to approach it from a Price of Non-Conformance perspective (including your 95% yield data) and feed that back into the engineering/technical community. I feel the dollar figures will help them prioritize their targets.

[The Fixer]

Marc,
Thanks for pointing me to the cost of quality thread and thanks for this bulletin board. There is really no better way to improve quality perspectives than by drawing on the vast experience of others in the profession.

Glenn D.