Did QS-9000 Fail? If so, Why?

[Marc]

This was an e-mail I received:

> I am putting together a presentation for a client and wanted to provide
> some justification for what QS-9000 was supposed to achieve and hence show
> the reasons for the changes in TS 16949. All I can get from the standard is
> that it was supposed to reduce variation in the supply chain. When I was on
> the TS 16949 working group I heard stories about QS-9000 not fulfilling its
> dream - talk about firms not making the linkages between FMEA, SPC and
> process variation but I have nothing official that I can use. I have looked
> at your site but cannot locate anything relevant. Can you help?

The following is as I understand the events and reasoning. I have nothing 'official'. Bear in mind the following are my opinions and that QS-9000 was written about 10 years ago:

What was QS-9000 REALLY supposed to achieve?

1. Reduced customer audits of suppliers - Savings to the supplier by a reduction of audits and savings to the Big 3 by allowing them to reduce SQA personnel (and shifting the cost to the supplier - 3rd party audits paid for by the supplier, of course).
2. Bring a common set of certain systems standards to their customers. The word 'harmonzie' is often used.
3. Ensure suppliers are following requirements both internal and external.
4. Officially: To reduce variation in the supply chain.

The original intent of QS-9000 was to reduce or eliminate multiple supplier audits by the 'Big 3' and to set a 'standard' set of basic requirements including aspects such as design and development via the AIAG's APQP process. This process included additional aspects such as requirements for FMEAs and control plans, as well as several other improvement elements such as an effective nonconformance and corrective action system. The issue of interpretations came heavily into play as did the number of suppliers which did not have anyone in-house who understood many of the requirements. Many companies did not even have someone who really understood the basics of control plan development, FMEAs and such. This said, many companies had never even done SPC and again had no one internally who really understood SPC.

This was combined with many QS-9000 auditors who did not understand the workings of things like control plans themselves, not to mention interactions of the systems as a whole. I should mention here that many customers often had SQAs which, sad to say, did not understand any more than the auditors and their suppliers. I met one QS auditor who was a college prof who 'dropped out'. He had no manufacturing or other business experience. The audit went OK but he was obviously not an appropriate auditor - his understanding of many aspects was lacking including the intracies and applicability of FMEAs and control plans, tying in SPC and such. in some cases, such as where the company had little internal understanding, it was a lot like the old 'blind leading the blind'.

Another reality was the auto makers are really product oriented. Even with QS-9000, customer audits at many companies did not subside as predicated. I was working with Motorola when Ford sent a team in for several days. While QS-9000 was nice to have, Ford really wanted product audits and Q1. In conversations there at the time, it was evident that the number of customer audits did not decline. Admittedly this is anecdotal, but I believe it was real.

What occurred was a serious learning curve for a lot of companies as well as the recognition that while the idea of a common set of automotive quality standards was and is a good idea, not every company expected the same things. The confusion was severe as evidenced by the flurry of interpretations and explanations of the document.

Another failure mode was that QS-9000 was developed mainly with high volume precision manufacturers in mind. This did not stand up as QS-9000 meant different things to manufacturers of different products. For example, transmission parts will have many critical dimensions and tight tolerances while a rubber grommet is quite a bit less. On the other hand, 50 grommets may go in each car in a number of models while for the dashboard manufacturer there was only one per car and may only be used on one platform.

Unfortunately, the document was too specific to be easily understood and applicable to all manufacturers.

What were the Good points about QS-9000?

QS-9000 did provide a good base for a quality system which went into enough technically that if implemented, including (for many, many companies this was problematic) having trained personnel, could not help but to have a positive effect on a company. In the 1980's I became involved in quality assurance in a military manufacturing environment. Most of the QS-9000 requirements were addressed because of the nature of the product - high reliability military electronics for aerospace and marine environments. From this experience I recognized that elements of QS-9000 were good, basic requirements for reliable products. From FMEAs to control plans, none of this was new.

What was new was applying these in a typical company. It was expensive ($500 hammers and toilet seats come to mind). However, again I think it is important to recognize the 'good' aspects of QS-9000.

Why ISO/TS 16949?

About 1998 I was privy in a meeting where what is now TS 16949 was being formulated. There were a number of issues including those addressed above, but in addition the 'Big Three' were recognizing they each had different things they wanted. QS-9000 started out as a 'mutual consortium' of 3 companies. By 1998, as often happens to mutual partnerships in projects between companies, the 'Big Three' wanted to offload the responsibility for the document and end this 'special' relationship.

In addition, there was a lot of pressure from mega companies like Motorola for significant changes. To some extent they extended their influence into the document content and such just as lobbyists do in politics. And remember - Some mega companies even refused to consider QS-9000 registration - Intel is an (admittedly rare) example of this.

Another aspect was that Europe watched as QS-9000 developed and spread. I generally see the following stated as the overall reasoning:

"16949 harmonizes QS 9000 with European standards for automotive supplier quality systems (VDA, ANFIA and CCFA/FIEV). ISO/TS 16949 has ISO 9001:1994 at its core and also incorporates some of the anticipated changes in ISO 9001:2000. These include increased focus on product realization and resource management.

I do not deny this is a large part of the TS movement. However, remember that the Japanese and most Pacific Rim auto manufacturers have not embraced TS 16949 any more than they did QS-9000.

What about the 'Process Approach'?

I won't expound on this much because I see it as silly from the view point of a biologist, which I am by training. Everything is a system so to me it's a 'no brainer'. A company has a billing system of one sort or another which HAS to interact with other parts of the company to function. And if billing does not function, the company cannot exist no differently than a body can live without critical organs.

Summary

In the long run, I'm not so sure it's an issue of QS-9000 failing. I didn't like QS-9000 personally as I felt it was extremely poorly written. But - taking into account the evolution of manufacturing capability and technology in general, it served a purpose, abit an expensive one for many companies. I see ISO 9000 in the same way with the exception that I think the 2000 revision was a poor one where the intent of ISO 9000 changed and in fact absorbed some qs-9000 requirements.

OK, folks. Now it's YOUR turn. Did QS-9000 fail in its objectives? If you believe it did, please take a minute and explain why you think it did or didn't.

If you have any 'official' explaination(s) about Why Qs-9000 was written, Why it failed and/or Why TS 16949, let us know about it/them!

 

[Howard Atkins]

Marc that was very well put
I would like to add some comments of my own.

I think that one of the reasons that QS9000 was implemented was as method of introducing and implementing into the manufacturers, especially the smaller ones an organized quality system.

As can be seen from the attached table (iso figures) which is from the "The ISO Survey of ISO 9000" page 2 the ISO 9000 standard was at the time of the publication of QS9000, and is still to a large degree a European standard. The Big 3 tried to expand the use of an agreed structure.
The fact that the US suppliers are not committed to ISO 9001 can be seen in the fact that a number of companies did not transit to ISO 9001:2000 from QS9000, which most Europeans did and allowed their ISO 9001:1994 registration to lapse while waiting to implement ISO/TS 16949:2002.

As Marc said

” Another reality was the auto makers are really product oriented.”

This is correct and the format of the ISO 9001:1994 was product orientated. The QS 9000 tried to add some more system approach ideas, such as: use of company data which the companies based on the product approach did not fully utilize, despite the various indicators required by Ford for example.

The structure of the QS9000 was also very bureaucratic, the insistence of the standard that all requirements be mentioned in the documentation makes me believe that the desk audit was more important that the actual implementation. This insistence that it be written down became more important than if it was actually performed. This fits in also with my impression that in a lot of cases the documents are more important that the reality. The fact that the some companies submit PPAP’s with the documents arranged in the order of the PPAP manual and place the PSW in section 13 as the manual lists illustrates this; the PSW in my opinion is the title page of the PPAP and gives an initial overview of the document. Some companies even add section 15 “Bulk Material Requirements Checklist (for bulk material PPAP only) as a page, so no one can say that they did not follow the manual. This is absurd and a total waste of time, effort and paper.
This might be approved manner but to me shows a misunderstanding of the process and the reason for the PPAP.

The demands for SPC were also based on a misunderstanding of the strengths and weaknesses of SPC.

“ SPC stands for statistical process control. It is unfortunate that in North America statistical methods are so routinely applied to parts, rather than processes.”

Page 2 AIAG SPC Manual.
The requirements of customers for SPC on dimensions whilst reasonable in some industries are nonsense in others. Plastic for example has a period of shrinkage, which can take 24 hours to reach 99% of the final dimension. The requirement to perform SPC on KPC’s during production is meaningless as the whole point of SPC is to have an online picture of the production and to enable action as a result of any drift that occurs. There is no point in doing this 24 hours after the production. When this is explained most SQE’s understand and will accept that this is not necessary but I feel that in a lot of cases this is not understood by the supplier and it continues with meaningless data. I have asked suppliers to chart dimensions and they have not queried it.

These issues of blind obedience to requirements is in my view counterproductive to a quality system as it does not allow any intelligent analysis of processes and in fact can cause the very problems that it is supposed to prevent.

Another problem issue in the approach of the QS9000 which has carried over into the TS is the subject of supplier development.
Despite the big words on partnership between suppliers and customers, the requirements are those of registration to standards and there is no reference to building a partnership. As Marc said

“Many companies did not even have someone who really understood the basics of control plan development, FMEAs and such.”

The customers themselves did not try to develop the suppliers in terms of educating them.
The development was that of carrot and stick, with a heavy use of the stick. I asked a number of OEM’s if they had schemes for improving suppliers and they could not offer anything. The courses offered are usually about how to fill in their forms.
I must add here that Delphi did have a Picos workshop programme and I worked in a company where 3 people came for 3 days to perform a reduction of set up workshop. This though is the exception.

One of the reasons that I heard about for the change was the fact that the multitude of registrars meant that there was unevenness in the application of the standard and as a result the TS registrars are much more regulated. There is in fact a closed shop.
Another aspect for the demise of the QS9000 as I see it is the fact that the Big 3 felt that they were loosing control of their suppliers, as one SQE from Ford said,

“I can take Q1 from you”

Thus Ford now wants Q1.

I have seen thought that the TS auditors seem to be taking the same administrative approach as the QS9000 and I am not sure that the spirit of the standard is being audited and not the letter.

The old problem of I want your company to look like mine because that is what I know!

Can a standard such as ISO/TS 16949 or QS9000 really be universal and ignore cultural differences.
I don’t know

 

[Wes Bucey]

Marc that was very well put
I would like to add some comments of my own.

I think that one of the reasons that QS9000 was implemented was as method of introducing and implementing into the manufacturers, especially the smaller ones an organized quality system.

These issues of blind obedience to requirements is in my view counterproductive to a quality system as it does not allow any intelligent analysis of processes and in fact can cause the very problems that it is supposed to prevent.

The old problem of I want your company to look like mine because that is what I know!

Can a standard such as ISO/TS 16949 or QS9000 really be universal and ignore cultural differences.
I don’t know

If my echo of Marc and Howard helps -
I feel the stated purposes of QS9000, just like the stated purposes of ISO9k2k, TS16948, AS9100, etc., are all excellent.

The breakdown comes in execution of the so-called Standard.
Customers want to reduce the internal cost of supplier research (deciding who is a good supplier) and think they can standardize the decisions in selecting a supplier.

Unthinkingly, they want "average" suppliers. When they say they want above-average suppliers, what they really mean is they want "super average" suppliers who complete the Standard requirements to a "T"

In the process of selecting super average suppliers, they squelch innovation and end up with a cadre of "me too, only cheaper" sycophants as a supplier base and thus doom themselves to mediocrity.

 

[Sam]

I do not believe QS 9000, as a quality management system, failed. What I believe failed was the "image" that QS came to project. The firestone caper and the Perry Johnson incidents, ill trained auditors, hundreds of registrars,etc.all helped to project an image that QS was a failure.
Hence, TS2, with a sharper image and more focus on the process. And only 53 registrars worldwide tightly controlled by one source; The IATF.

 

[Marc]

Delphi's PICOS

...I must add here that Delphi did have a Picos workshop programme...

I went through PICOS when I worked with Delphi on the airbag project years ago. A good workshop.

The old problem of I want your company to look like mine because that is what I know!

This was always my biggest basic complaint about QS-9000.

Thus Ford now wants Q1.

As I remember, Ford never formally required QS-9000 registration.

 

[Sidney Vianna]

My 2 centavos

I believe that it failed for many reasons. The most serious and structural problem is the fact that the American Automotive Industry is intrinsically opposite to one of the ISO 9000 Principles: Mutually Beneficial Relationships. With one exception, American Automotive OEMs want to squeeze their suppliers as much as possible. There is no trust between customer and suppliers and, for the most part, the Procurement Functions, at the customer organizations, supersede Quality, and focus solely on price reduction, without paying attention to value.

Suppliers always resisted to the mandate for Registration, and, for the most part, after procrastination, many took the path of least resistance, i.e., the registrars/auditors that made the registration mandate less “painful”, was hired. Because of the price pressure, many automotive suppliers took the same approach as it came to registrar selection. Whoever was the cheapest got the contract. As we know very well, when you add shoddy organizations with incompetent auditors, you get the literal what you pay for. Due to price war, amongst Registrars, many certification contracts were signed with no profit margin. Guess what happens when the registrars operate at no profit? Outsource audits to cheap assessors who free lance. Don’t train nor develop your auditors, leading to even worse, ineffective audits. Thanks to my senior management that made a conscious decision of never signing a contract at a financial loss. Helps keep our profits in order, and supports our integrity.

With the accountability for effective QMS implementation and assessment dispersed between customer, suppliers, trainers, consultants, auditors, accreditation bodies, etc. . . nobody is truly accountable for ineffective supplier QMS. How many horror stories we heard of? All the loopholes were explored. When you have to define the minimum number of hours that constitute an auditor-day, you know that you have non-serious players in the process. Instead of disapproving registrars that were blatantly circumventing the rules, more interpretations were invented. But nobody was ever publicly punished, i.e., lost their accreditation, permanently. Until seriousness of the process is established, the cheaters are rewarded.

The fact that the Standard is very prescriptive and the interpretations process not always straight forward did not help either. Remember when we had to issue interpretations to previously sanctioned interpretations?

 

[The Taz!]

All points well taken. . .

I can see value in all of the opinions and statements offered. .

Having been on the implementation of end of Mil I, Mil Q, ISO, QS, and now TS I have watched an evolution in the standards.

I believe that if my info was correct, ISO grew out of our own quality systems from WW II. I believe that the BS standards (British Standards for the neophyte. . . Pamphlets really) were generated to start a European (Ally) movement to duplicate / assimilate the US quality systems that helped produce high quality ordnance at such a high volume and fast pace.

BS grew into ISO. . . I remember designing product to BS standards and holding those little pamphlets in my hand and thinking . . ."wow. . . we do this anyway"

What I see has happened, from having worked in one of those companies that did such a fine job and help set the pace and standardize the fastener industry (Held those ships, tanks, guns, planes, etc. together) . . . is that the "original intent" has been lost in standardizing quality systems. What started out as a legitimate attempt to bring the allies along. . . has turned into a financial bonanza for some registrars and Certifying bodies. . . afterall, they ARE businesses.

Our own systems are coming back at us. . . and the interpretation of the "original intent" and focus has changed.

FMEA's were first developed in the 60's by aerospace engineers. Adoped by the automotive companies. . . a very useful (If done correctly) and very misunderstood tool. All companies that I have had the "priviledge" of working with look on them as a required task and a pain in the neck. . not a tool. The PFMEA is one of two documents I want in my back pocket when I go on the production floor troubleshoot a production or quality problem. If the operator on the floor were trained (Basically) in interpreting a PFMEA, a company might meet the requirement for "Awareness" (How their job effects the customer and how the concequences of poor quality effect the customer) in TS.

What I observed was the PFMEA (and DFMEA) being looked on as "new" documents and practices. This was primarily by "new" management, manufacturing, and engineering personnel. It ain't new gang. . .

I may sound off-topic here. . . but what I think I am trying to say is that ISO, QS, and now TS ain't new kids on the block. I think the international community is attempting to re-package the basic "good practice" of quality and manufacturing itself.

It is well documented that we, as a country, produced a heck of a lot of goods after WW II. . What we made was sold. . . good or bad. . . that set a precident for poor quality becomning the norm (Generalization). "If you make it. . . They will buy it" as opposed to "If you make it right. . . you will prosper". . .

I personally agree with the Process Approach of TS2. . . it reeks of Deming. . . I have found it easier to implement primarily due to less documentation requirements. . . only 6 required procedures and the emphasis on "Experience and training" being a recognized measure of competence. . . . If you look at the required "procedures," they form the basis of what I believe the "original intent" was. . . Training, Control of the goofs (Don't send them to the front lines), and fix the goofs, and Documents (Rev control being the operative part). . . in short. . .Plan-Do-Check-Act.

Audits DO focus on auditing systems and processes . . . not people and "requirements". . . 20 disjointed requirements (Elements), without interraction are black and white auditable. . . process auditing requires the auditor to look at the "process" and look at meeting "the intent" of the standard.

IMHO, QS didn't fail. . . it is a step in an evolutionary process back to our quality roots.

Enough of my rambling. . .

Ciao!

 

[Marc]

Well, yes - QS-9000 is a component of an evolutionary process. But like looking at human evolution, Neanderthal man was one 'branch' which we single out to study and ask "Why did that branch die out?"

If QS-9000 was a 'good' document, it would not have been replaced in under 10 years from its release. Rather, the document its self would have evolved, as ISO 9001 has, rather than phased out (or how ever you want to describe its coming obsolescence). Anthropologically speaking, I guess we can say QS-9000 failed to survive.

 

[Laura M]

Well, yes - QS-9000 is a component of an evolutionary process. But like looking at human evolution, Neanderthal man was one 'branch' which we single out to study and ask "Why did that branch die out?"

If QS-9000 was a 'good' document, it would not have been replaced in under 10 years from its release. Rather, the document its self would have evolved, as ISO 9001 has, rather than phased out (or how ever you want to describe its coming obsolescence). Anthropologically speaking, I guess we can say QS-9000 failed to survive.

Why aren't you considering TS an evolution of QS? Would TS have come about without the B3 starting QS first? One of the reason QS became obsolete was that the foreign companies "liked it" but didn't have any ownership so they needed an independent document, right? Not sure the significance of 10 years to be a 'good' document.

 

[Marc]

TS is an evolutionary off shoot, but it is not QS. I will agree it is a mixture of ISO 9000 (now ISO 9001:2000) and QS along with some stuff from VDA, etc., but QS-9000 is dead for all intents and purposes.

As to 10 years, I was liberal. 1995 to 2000, when TS really came out, is 5 years. And considering that TS was in the works by 1998 - that's about 3 years. It's failure to achieve it's goal was evident early. In 1998 I was in a confidential meeting where this was discussed and TS 16949 was being 'planned'.

You may not define it as a failed 'standard'. I do.