Paradigm Shift in QA? Another View

B

Bill Pflanz

I wouldn't be too harsh on Six Sigma. In all honesty, you could have put TQM, ISO 9000 or Lean Manufacturing or other quality "fixes" in replace of Six Sigma and it would not have read any differently. Once Six Sigma gets totally tossed out, something else will replace it.

The interesting part is that the Cove is filled with reputable quality practitioners who use the same tried and true methods no matter what it is currently called. It is the passing of those methods and concepts to the next generation that is probably the most important event that must occur.

Bill Pflanz
 

Jim Wynne

Leader
Admin
I wouldn't be too harsh on Six Sigma.

I was being harsh on the people who use SS as a smoke screen, and as a substitute for doing some real work to improve things. It does us no good to get harsh with inanimate objects.

In all honesty, you could have put TQM, ISO 9000 or Lean Manufacturing or other quality "fixes" in replace of Six Sigma and it would not have read any differently. Once Six Sigma gets totally tossed out, something else will replace it.

I have no idea what "TQM" means, and neither does anyone else, because it means only what the person talking about it wants it to mean. Do you think this sort of thing will go on forever? How long are you willing to have your knowledge, experience and integrity devalued?

The interesting part is that the Cove is filled with reputable quality practitioners who use the same tried and true methods no matter what it is currently called. It is the passing of those methods and concepts to the next generation that is probably the most important event that must occur.

I think the time for that sort of passivity is gone and the time for righteous indignation is upon us.
 
B

Bill Pflanz

I was being harsh on the people who use SS as a smoke screen, and as a substitute for doing some real work to improve things. It does us no good to get harsh with inanimate objects.

Okay then don't to be too hard on the SS users. They are no different than people using prior improvement methods with their associated buzzwords.


I have no idea what "TQM" means, and neither does anyone else, because it means only what the person talking about it wants it to mean. Do you think this sort of thing will go on forever? How long are you willing to have your knowledge, experience and integrity devalued?

That is my point. How many discussions have there been about the meaning and value of SS or Lean or any other quality fad term? It will go on as long as someone wants to sell a new book or seminar to get the attention of management. What is constantly being devalued is management knowledge, experience and integrity who buy the latest fad.

I think the time for that sort of passivity is gone and the time for righteous indignation is upon us.

Most of the people in the Cove do what they can here and in their jobs. There is not much we can do about the latest fad but not participate. You and I have both been around long enough to know that we do the best we can when management jumps at the latest quick fix.

Bill Pflanz
 
Q

QCAce

A different viewpoint….is a QA paradigm shift needed?….yes

I like the article presented. But what it tells me is not whether SS is leading us down a wrong path, I believe we were already on that path.

My young frame of reference (QA worklife-wise), only takes me back to the launch of ISO9000.

The pride of ISO was quality improvement via registration to a standard, (the more customers and governments requiring ISO, the more companies would register and the more quality would naturally improve). But as we soon (or not so soon) learned, poor companies shipping out poor product could still achieve ISO registration, even with poor results.

The pride of TQM, (and the related off-shoots thereof), was more focus on results. This was a step in the right direction, but the question remains as to what are the right results to focus on?

The pride of SS was to put some focus onto some of those results, namely DPM and $$ saved. A noble direction to take. Similar to the credibility doubts surrounding ISO registration, however, requirements for SS certifications also raise questions. And as the article points out, there are still questions raised as to whether the results of certain SS projects are results worth focusing on to begin with?

The trends I see repeating are the same issues every new quality professional faces when they come to company and are required to “develop and implement the quality management system.” Most of us are engineers, which only makes things worse because as engineers we like to put things together in terms of A – B – C:
A = build the quality system basics, (doc control, corrective action, ncmr, internal auditing, etc.)
B = fill in the system for each department, (procedures, work instructions and records)
C = monitor the results and improve

During the ISO period, the results we would focus on would be things like, “number of work instructions found in use that were not to the correct revision,” and all sorts of advanced document approval systems would be launched while poor quality product was being shipped.

During the TQM phase, improved focus on profits and costs were analyzed, but the hidden costs remained hidden, (the bottom of the iceburg).

Now with SS, I’ll hear statements like, “for my blackbelt certification project I improved the fabrication process of part Y of subcomponent Z to a Cpk of 2.5, saving the company $X of scrap and rework per year.” In the meantime, without knowing that the poor quality of their bread-and-butter, product A, was driving them out of business.

The paradigm shift I’m trying to implement is to build the system as C – B – A. The results monitoring comes first and is the foundation. The continuous improvement cycle is then setup in each department. Capping it off with some basic quality system processes.
 
The QA pros that I have learned from have never touted the latest fad, buzzwords or made any claims other than, "If you don't perceive that I have made things better for this company, fire me."

Personally, I try to be guided by Deming's Out of the Crisis as a whole work and focus on a big picture rather than just "Quality." I am an engineer with enough economics and accounting training that this is possible.
1) Establish some kind of system to measure where your gains and losses are. Not just stuff like scrap and rework on high volume products but also engineering projects that spend money but don't develop anything marketable, salespeople who spend money visiting and wooing customers that don't buy anything, accountants that hold engineer's and salespeople's feet to the fire for tiny expenses without noting the huge returns they achieved, etc., ad nauseum.
2) Do something productive with the data. As mentioned above, reducing the scrap rate from 50% to 0% on a product that sells 2 pieces a year for $1,000 and costs $500 to make is not nearly so important as reducing the warranty return rate from 0.5% to 0.25% on a product that ships 1,000 pieces a year for a $1 each costing $0.50 to make, but that incur a $250 charge back from your customer, an 8D report and a plant wide containment action for each return.
3) Know when you are done and move on. I had a manufacturing V.P. that insisted that we 100% test PCB's before installing them in units because the first pass yield on the final assemblies was only 87.5%:mg: Unfortunately, there were 6 PCB's in every unit and each board took 30 minutes to test as an individual assembly. The final unit took 1 hour to assemble and 1 hour to test. So a good tech could test, troubleshoot, reassemble and test 1 bad unit in about 3 hours (remember that number). Our monthly production was sixteen units (1 defective PCB out of 96 or ~1% = 1 defective unit out of 16 or 12%). Our monthly testing bill to achieve 100% FPY on final = 48 technician hours. Our monthly testing bill to achieve 87.5% FPY on final but still ship 100% good product = 3 technician hours (we had to spend 16 hours testing the units either way so as to establish traceability and issue calibration certificates). BTW, the 6 boards combined contained over 12,000 opportunities for defects when parts counts, solder joints, nodes, wiring connections, etc. were taken into account. I had been tracking defects for 12 months and seen 12 different defects.

Any real QA person would go look for more verdant pastures. Can you imagine the expensive engineering time needed to weed out and eliminate 12 discrete defects, over 12 months, 16 units a month each with 12,000 opportunities (i.e. 12/2,304,000 or 5.2 PPM)? Instead, I spend about 80 hours designing, automating, testing and error-proofing a database to hold both calibrating equipment data AND calibration data for our final products. Upon completion of test, a technician would log in, pull down a menu and fill in the Model and Serial Number of a product and all of the data for a cal cert would fill in automatically and with a push of a button, cal cert for each and every product. Time saved every year in order entry, production, manufacturing engineering, shipping, etc, over 1,000 hours.

Conclusion: Our customers were far more excited about receiving an iron-clad, audit-proof calibration certificate with every product than they would ever be about our 100% FPY braggadocio. For you other kind of QA people: keep touting SS, quoting impressive but empty statistics and waving your certifications. I'm making a fortune saving real companies real money and I'll still be able to do it 10 years from now when the newest QA fad is but a memory.
 
Last edited:
B

Bill Pflanz

Andy and Icy seem to have found a way to get the job done without the latest fad or buzzwords. Nice job. :applause: As I said before, many of our Covers are getting it done - sometimes in spite of management.

Bill Pflanz
 
J

JaneB

The paradigm shift I’m trying to implement is to build the system as C – B – A. The results monitoring comes first and is the foundation. The continuous improvement cycle is then setup in each department. Capping it off with some basic quality system processes.

Agree to some extent - but surely any 'system basics' (A) must be built upon, & stem from, the organisation's strategy & objectives in the first place?? In other words, the place to start for anyone (engineer or not, I'm for equality of opportunity & non-discrimination ;) is from the top down, not the bottom up. Otherwise, it's like a civil engineer say, starting with building & then working back to the drawings... or as IT people like to parody, 'never mind the functional specifications, let's just get to writing the code & sorting it out later'.

One of the things that bothered me about the article was the exclusion of anything that wasn't part of actual production. eg:
ZERO: That's Marketing, not Production. ...In the same way, quality management is just one piece of the puzzle, but not the answer to the whole puzzle."
or
ZERO: If you people in Marketing can't figure out what products we should make better, then there's the problem.

While that might be a reasonable reflection of an SS approach, it doesn't in way reflect my understanding of a real quality management system. This kind of silo thinking (we're making perfect product, it's all Marketing's fault) is lunacy. Not to mention business suicide.
 
D

David Hartman

I believe that it was Dr. Deming that stated that we (businesses) need to constantly and forever be improving. Many of us strive to accomplish this (as the article implies) by using our Quality tools (SS, SPC, etc.) to minimize variations / defects in our products, services, and processes.

But is there more to it? Shouldn't we (as a company) be looking at how to continually provide the customer with increased value, by constantly striving to improve our products and/or product line so that we are always providing them with something unexpected. It is a risky venture, who among us remembers the Cadillac V8/6/4? But, this continues to be a practice that sets many a Japanese car apart from the rest.

Quality is NOT a department. It should never be departmentalized. It must be the function of every department within an organization. Yes, we need to continue to practice the activities defined by classical industrial quality control, but we must also strive as a company to do more than meet the customer's requirements. Using Tom Peter's words, we have to WOW them.
 
Q

QCAce

Agree to some extent - but surely any 'system basics' (A) must be built upon, & stem from, the organisation's strategy & objectives in the first place?? In other words, the place to start for anyone (engineer or not, I'm for equality of opportunity & non-discrimination ;) is from the top down, not the bottom up. Otherwise, it's like a civil engineer say, starting with building & then working back to the drawings... or as IT people like to parody, 'never mind the functional specifications, let's just get to writing the code & sorting it out later'.

Okay......I would not disagree....
But what I would say is try thinking of the Metrics as the drawings for your building or the specifications for your IT code.
:yes:
 
Top Bottom