Paragraph 7.3.2.3 Special Characteristics,of ISO/TS16949 states: "The organization shall:argue: :argue: identify special characteristics" The customer has not identified any special characteristics on the product we are producing. The auditor stated that we needed to identify a special characteristic even though the customer did not.

This runs against my understanding of what a special characteristic is to be used for. Is the auditor right?

This runs against my understanding of what a special characteristic is to be used for. Is the auditor right?I agree with you. In my understanding if Eaton did not identify any special characteristic for your products, nor did your organization, then there should be no special characteristic. Period. To artificially identify a characteristic as special just satisfy an external auditor is non value added.

I also agree. YOUR customer dictates what is a "special Char." not you. I did find that being a "smaller company" (under 50 employees) we did a little "bending" in order to become certified. After the certification, alot of "stuff the auditor wanted was deemed "Non-value-added" and dropped. Our auditor wanted us to show a bar graph to show trends with customer rejections and on-time delivery. We had 2 customer rejects in december and non all year otherwise. we also had a 100% otd across the board. Well to make him happy, we made bar graphs showing 1 bar rising to "2" on the scale for customer rejects and a flat 100% bar across the graph for otd. Hey it is whatever it takes sometimes.

:)

:agree1:

the auditor has nothing to say with this.

its the customer who says which is a CTQ(critical to quality)/special char.
if needed the organisation will decide upon the CTQ based on their experience on that product or other group of products.

the auditor is trying to act smart and dictate you.

shesha

Paragraph 7.3.2.3 Special Characteristics,of ISO/TS16949 states: "The organization shall:argue: :argue: identify special characteristics" The customer has not identified any special characteristics on the product we are producing. The auditor stated that we needed to identify a special characteristic even though the customer did not.

This runs against my understanding of what a special characteristic is to be used for. Is the auditor right?


I agree with Sidney, but think the truth is somewhere in the middle. The customer determines "Customer" dictated special characterisitcs. Even in QS, the APQP or FMEA manuals encouraged "organization" designated SC's. TS just brought it into the standard itself.

I would suggest the intent here is that organizations never selected SCs, even though often they clearly had certain characteristics that caused problems. When such situations exist, then an organization should designate these problem characteristics as organization designated SC's.

I would suggest the intent here is that organizations never selected SCs, even though often they clearly had certain characteristics that caused problems. In the litigious society we live in, well managed organizations would be the primary interested parties to minimize failures that could lead to unsafe situations. Thus, if they know what they are doing, they would identify and designate special characteristics, when warranted, and, apply the appropriate levels of control when dealing with such special characteristics. It should all be part of managing their risks. Sometimes, an ill advised auditor will demand corrective actions without considering the economical consequences of his/her write up.
I can envision the following dialogue

Auditor: Does your customer spec identify any sc?
Auditee: No
Auditor: Did you identify any sc?
Auditee: No
Auditor: But you have to.
Auditee: Why?
Auditor: Because it is in the Standard.
Auditee: But what characteristic do you think should be a sc?
Auditor: I don't know. It is your product. You should know it.
Auditee: And we do. We reviewed the design of the product and did FMEAs. We did not identify any sc.
Auditor: But you need to. It is in the Standard...
Auditee::frust:

In the litigious society we live in, well managed organizations would be the primary interested parties to minimize failures that could lead to unsafe situations. Thus, if they know what they are doing, they would identify and designate special characteristics, when warranted, and, apply the appropriate levels of control when dealing with such special characteristics. It should all be part of managing their risks. Sometimes, an ill advised auditor will demand corrective actions without considering the economical consequences of his/her write up.



I agree with you. It is one thing for an auditor to point out that a particular problem part would benefit from an SC. It is way different and wholly inappropriate for an auditor to expect every part to have one and start throwing his weight around.

I think it goes to the intent of the requirement, and this auditor appears to be way over the line.

This may not make it past all auditors, but it does work for many of them.

1) Establish clear criteria for establishing a special characteristic.
2) Establish a review process for special characteristics (probably part of a design review)
3) Document the review, including a formal statement (if applicable) that no characteristic met the criteria for a special characteristic.

Most auditors are more concerned that the process exists and is used correctly, than in a special characteristic for every product. Although there are always a few that are.

...
Most auditors are more concerned that the process exists and is used correctly, than in a special characteristic for every product. Although there are always a few that are.


Exactly. An auditor should not insist every product have a special characteristic. But it is hard to believe no product needs one either. Particularly, when you observe piles of nc product and customer complaints and lots of special inspections.

These are all indicators of problematic characteristics that are not adequately controlled.

I agree with you. In my understanding if Eaton did not identify any special characteristic for your products, nor did your organization, then there should be no special characteristic. Period. To artificially identify a characteristic as special just satisfy an external auditor is non value added.Sidney,
For the first time, I disagree with you. Talk to your auditors. They are requiring organizations to identify special characteristics. Full stop.
The organization SHALL identify special characteristics [see 7.3.3 d)] and....d) specify the characteristics of the product that are essential for its safe and proper useI had this debate with my DNV auditor. It is a non-starter. Since our reason for being is to supply electrical power, we don't really consider its supply a "special characteristic". However, we measure and control the voltage and current output of every unit because those measurements are ESSENTIAL to the safe and proper use of the unit.
Everyone has:
Product characteristic or manufacturing process parameter which can affect safety or compliance with regulations, fit, function performance or subsequent processing of the product.
Step out of the forest, identify the essential trees, and add the little squiggly § to your FMEAs, Control Plans, and Work Instructions. I fought this one, too, until I realized that there is a lot of value to identifying the ESSENTIAL characteristics of our products that we had taken for granted.
-Icy

Sidney,
For the first time, I disagree with you. Talk to your auditors. They are requiring organizations to identify special characteristics. Full stop.
I had this debate with my DNV auditor. It is a non-starter. Since our reason for being is to supply electrical power, we don't really consider its supply a "special characteristic". However, we measure and control the voltage and current output of every unit because those measurements are ESSENTIAL to the safe and proper use of the unit.
Everyone has:

Step out of the forest, identify the essential trees, and add the little squiggly § to your FMEAs, Control Plans, and Work Instructions. I fought this one, too, until I realized that there is a lot of value to identifying the ESSENTIAL characteristics of our products that we had taken for granted.
-Icy

I'm going to come down on the other side of this one, and say that the interpretation that there must be special characteristics is too invasive. My interpretation of "...shall identify special characteristics..." is that if there are any special characteristics, they must be properly identified as such, and I (or my customers) will decide whether there are any or not, not an outsider.

No matter what product or service you provide, there are certain characteristics that are "special" in the sense that they are the most important to the parties affected by the product or service. An oragnization benefits from identifying them (even if the customer does not identify them) by directing resources to eliminate or mitigate the largest risks - failing to provide those special characteristics. Reducing the largest risks first is the most economical way to proceed - all else is waste.

Thankyou all for a very good and informative discussion. My problem is what can be a special characteristic if our process is extremely simple and has produced on output that is rated at 73 defects per million over the past 3 years?

Our process is to make a harness that has 5 wires, from two electrical components, attached to a connector through a corregated tube. There is a 100% electrical test performed to validate the output required for correct operation.

We do have a completed and operational Process FMEA and Control Plan with identified controls for each of the steps in our process that obviously controls our process sucessfully. What advantage is there to identify a SC on the FMEA and Control Plan?:thanks:

Thankyou all for a very good and informative discussion. My problem is what can be a special characteristic if our process is extremely simple and has produced on output that is rated at 73 defects per million over the past 3 years?

Our process is to make a harness that has 5 wires, from two electrical components, attached to a connector through a corregated tube. There is a 100% electrical test performed to validate the output required for correct operation.

We do have a completed and operational Process FMEA and Control Plan with identified controls for each of the steps in our process that obviously controls our process sucessfully. What advantage is there to identify a SC on the FMEA and Control Plan?:thanks:

You basically have two choices at this point:

Talk the auditor out of his foolish demand
Use the dartboard method to choose and document a special characteristic so as to make the auditor happy.Which method you choose depends on how obstinate the auditor is and how willing you are to press the issue.

No matter what product or service you provide, there are certain characteristics that are "special" in the sense that they are the most important to the parties affected by the product or service. ISO TS 16949 has a definition for special characteristics. It does not jive with your description above. If we do not stick to issue at hand and start thinking of "special" characteristics, when the Specification talks about "special characteristics", we will not go anywhere.

Step out of the forest, identify the essential trees, and add the little squiggly § to your FMEAs, Control Plans, and Work Instructions. I fought this one, too, until I realized that there is a lot of value to identifying the ESSENTIAL characteristics of our products that we had taken for granted.
-IcyIcy, I agree with you that, in most cases there will be special characteristics, but there might be cases where it doesn't. And an auditor should not artificially impose his/her will onto the organization they are assessing. In your case, it looks like you gave consideration to the product and agreed that sc's existed.

If I supply a commercial washer that is only used to secure the license plate of a car to the body, I might not have any special characteristic (imho). Remember when you designate a sc, you are expected to enhance the controls over it. What many times impacts it's costs. We all know how sensitive the automotive supply chain is to costs. I truly believe that sc is a component of the product safety/liability risk management process and it is possible for a very simple, non critical component not to have a sc. At least imo.
Thanks for your comment.

Our process is to make a harness that has 5 wires, from two electrical components, attached to a connector through a corregated tube. There is a 100% electrical test performed to validate the output required for correct operation.

We do have a completed and operational Process FMEA and Control Plan with identified controls for each of the steps in our process that obviously controls our process sucessfully. What advantage is there to identify a SC on the FMEA and Control Plan?There you go. 100% Electrical Test for validation of output for correct operation. Special Characteristic.

The advantage is communicating this through your system so that everyone understands why you do 100% electrical test.
Icy, I agree with you that, in most cases there will be special characteristics, but there might be cases where it doesn't. And an auditor should not artificially impose his/her will onto the organization they are assessing. In your case, it looks like you gave consideration to the product and agreed that sc's existed.

If I supply a commercial washer that is only used to secure the license plate of a car to the body, I might not have any special characteristic (imho). Remember when you designate a sc, you are expected to enhance the controls over it. What many times impacts it's costs. We all know how sensitive the automotive supply chain is to costs. I truly believe that sc is a component of the product safety/liability risk management process and it is possible for a very simple, non critical component not to have a sc. At least imo.
Thanks for your comment.I don't think the auditor is imposing his will. I think that the standard is clear and the auditor is telling you that you have special characteristics that you need to document, without providing consulting services, of course.

I've replaced my share of license plates and I'll guarantee that the ID and OD of that washer are special characteristics. I don't think that you need "enhanced" controls. Just document what the controls are that keep those IDs and ODs in line. Are you saying you have no control over the dimensions of the washer? These are the arguments the TS auditors are throwing at us. I'm playing devil's advocate here; I don't like putting squigglies all over paperwork just because the standard says so, either. However, I see the point of the excercise. It's the same as Control Plans that are just a summary of the controls I already have in place. It's a communication tool. I do find some of these TS items tedious but there is an alternative: don't register.

There you go. 100% Electrical Test for validation of output for correct operation. Special Characteristic.

The advantage is communicating this through your system so that everyone understands why you do 100% electrical test.
I don't think the auditor is imposing his will. I think that the standard is clear and the auditor is telling you that you have special characteristics that you need to document, without providing consulting services, of course.

I've replaced my share of license plates and I'll guarantee that the ID and OD of that washer are special characteristics. I don't think that you need "enhanced" controls. Just document what the controls are that keep those IDs and ODs in line. Are you saying you have no control over the dimensions of the washer? These are the arguments the TS auditors are throwing at us. I'm playing devil's advocate here; I don't like putting squigglies all over paperwork just because the standard says so, either. However, I see the point of the excercise. It's the same as Control Plans that are just a summary of the controls I already have in place. It's a communication tool. I do find some of these TS items tedious but there is an alternative: don't register.

I agree on the electrical example. You may have a good PPM on returns, but what is your internal failure rate? The very fact that you have a 100% inspection on this means that you have already determined that it is special. Simply identify it as such and document the control that you already have in place.

The washer example in my mind becomes more of a judgement call. Yes, the ID is important from the standpoint of inserting the fastener, but will the customer really benefit from a reduction in variation? Or is in-spec good enough?

Thankyou all for a very good and informative discussion. My problem is what can be a special characteristic if our process is extremely simple and has produced on output that is rated at 73 defects per million over the past 3 years?

Our process is to make a harness that has 5 wires, from two electrical components, attached to a connector through a corregated tube. There is a 100% electrical test performed to validate the output required for correct operation.

We do have a completed and operational Process FMEA and Control Plan with identified controls for each of the steps in our process that obviously controls our process sucessfully. What advantage is there to identify a SC on the FMEA and Control Plan?:thanks:

Hi,

is 100% inspection a customer requirement.
if so it should be a special characteristics.
let it be a ISO/QS/AS system there is a reason for the customer demanding for 100% inpsection.
Special Characteristics are marked only to bring to the notice of the manufacturing plant that they are looking for a Cp/Cpk >1.33 or even more (depending on the customer).
if 100% inspection is not a customer requirement & if you have decided that you need to do 100% inspection then you are not confident abt ur process based on your past experience.
at this point of time based on your quality data & experience you can decide whether to make any parameter SC or not.

Apart from this I want to say that if the customer has not mentioned any SC's in his drawing & at the same time you too do not find any SC's
just simply mention that on the PFMEA that no SC found as Jim rightly said.

This should satisfy the auditor and will also ensure that this parameter is not overlooked by you.

Shesha

No matter what product or service you provide, there are certain characteristics that are "special" in the sense that they are the most important to the parties affected by the product or service. An oragnization benefits from identifying them (even if the customer does not identify them) by directing resources to eliminate or mitigate the largest risks - failing to provide those special characteristics. Reducing the largest risks first is the most economical way to proceed - all else is waste.


I agree, but as an auditor, I cannot require it on every product. There may in fact be some parts which do not have any particularly special characteristics. I think we have to evaluate on a case by case basis. The standard now clearly seems to lean in that direction, though. But, it has to be value-added.

ISO TS 16949 has a definition for special characteristics. It does not jive with your description above. If we do not stick to issue at hand and start thinking of "special" characteristics, when the Specification talks about "special characteristics", we will not go anywhere.

Sidney, I think we are arguing semantics...

ISO/TS 16949:
3.1.12
special characteristic

product characteristic or manufacturing process parameter which can affect safety or compliance with regulations, fit, function, performance or subsequent processing of product.

Spitler:
special characteristic
the most important characteristic to affected parties

If a product or service had no special characteristics, (according to the TS definition) it would be useless.

I think the auditior rightly requires that SCs be identified for each product or service so an organization directs its resources to the right places to enhance customer satisfaction (5.2).

Anything critical to the proper operation and safety...

Just be a good fellow and transpose your critical product performance specs and or dimensional specs into the control plan (which you probably do anyway) and call them "special characteristics" whew that was tough...and NO you don't have to use funky symbols as you have said the customer does not designate any. It may actually take 10 minutes worth of work.

Basically, I agree with Sidney. However, you could win the battle and lose the war. Is it worth the risk? You have to weigh the pros and cons to make a decision how you will be affected.

ISO/TS 16949:
3.1.12
special characteristic

product characteristic or manufacturing process parameter which can affect safety or compliance with regulations, fit, function, performance or subsequent processing of product.


Using this definition, all characteristics are "special," or at least potentially so. In my opinion, a great deal of time, money and effort is wasted on fretting over special characteristics, and the auditor's myopic vision in the subject of this thread is a perfect case in point.

There is never be any good reason to distinctly designate a characteristic as "special." It completely contradicts the intent and purpose of the whole APQP process, and presupposes that the people responsible for meeting specifications can't be trusted.

If the APQP process is faithfully followed, here's what should happen (after the design concept has been defined):

A basic design plan is devised, which includes definition of the purpose the design is intended to serve.
A basic feasibility review is done. Is it reasonably possible to manufacture and sell (at a reasonable profit) the thing in question? Are the necessary materials, skills and technologies available?
The DFMEA process is initiated. What needs to be taken into consideration in order for the design to be successful? What are the inherent risks, and how can they be mitigated?
A preliminary design is developed, and prototypes are built and tested. Changes are made based on testing and further review, and specifications are developed for fit, form, and performance.
Given the specifications, a manufacturing process is developed, and the PFMEA process is initiated.
Trial manufacturing runs are done, and process capability is evaluated.
Changes to specifications are made, if necessary, in light of lessons learned in the prototyping and trial run activities.
A production version of the specifications is established, which includes dimensional and performance tolerances.Note this is necessarily something of an oversimplification, and the steps listed above are not necessarily strictly chronological; some of them might happen concurrently.

Once specifications have been established, it should be expected that products that meet the specifications will serve the intended purpose, and it should further be expected that those responsible for manufacturing are aware that all specifications must be met. If we know in advance that the specifications are efficacious, and that producers are capable of meeting them, we should also be able to assume that the need to designate certain characteristics as "special" beyond the design stage has been obviated.

Designation of "special" characteristics on product drawings (or in other specifications) sends a clear message: "We don't trust our design and/or our suppliers." Fix those problems and you won't need special characteristics.

Thanks Jim, you made me realize something I have been missing. I should be correlating the Highest RPNs in my FMEAs to the critical (I refuse to use the word special in this post anymore), characteristics! As far as the argument goes. My "approach" is to maintain the certification through the least path of resistance, process owners have to be responsible for the quality their processes, not myself as the internal auditor, and sometimes there may be inherant conflicting goals as well as some suboptimization. (hence being an evil practioner) Ok so I like to babble ;)

Thanks Jim, you made me realize something I have been missing. I should be correlating the Highest RPNs in my FMEAs to the critical (I refuse to use the word special in this post anymore), characteristics!

You're welcome, but what I was saying was that risks should be taken into account in the basic specifications, and if the basic specifications are adhered to, there's no need for special/critical/super-duper characteristics to be identified anywhere past the design stage.

...Once specifications have been established, it should be expected that products that meet the specifications will serve the intended purpose, and it should further be expected that those responsible for manufacturing are aware that all specifications must be met. If we know in advance that the specifications are efficacious, and that producers are capable of meeting them, we should also be able to assume that the need to designate certain characteristics as "special" beyond the design stage has been obviated.



Darn Jim, I was with you all the way to the end, then the ball went foul...

BSpitler: I think Jim Wynne presents a very good explanation. Your logic makes sense, but if every part has to have special char., then they become routine and quickly cease to be special. If your point were correct, then we have to discard years of Big 3 practice. The custome has been to define troublesome, or particularly significant characteristics, or certain char. far more significant in a safety or fit situation. In other words, a characteristic that is special. If this were a sea change in their thinking, we would have heard something.

Jim, the only part I disagree with, is we don't live in a perfect world. There will be some occasions where a particular characteristic deserves to be called out for various reasons, to receive special. I "never" subscribe to a "never" or "always" approach. It generally doesn't happen. But I agree with you as to focusing on what matters.

There...now I've offended both sides...my work is done here...:D

I just think classifying characteristics based upon importance makes good sense because all characteristics are not created equal.
Tightly controlling the wrong characteristics will avail nothing. Controlling the right characteristics is everything when it comes to reducing risk and satisfying customers. Trivial many, Vital Few...

I just think classifying characteristics based upon importance makes good sense because all characteristics are not created equal.
Tightly controlling the wrong characteristics will avail nothing. Controlling the right characteristics is everything when it comes to reducing risk and satisfying customers. Trivial many, Vital Few...

Trivial few/Vital many is properly applied to expense caused by errors, which doesn't necessarily have anything to do with special characteristics. I'll say it again: If the specifications reflect what's really needed, and you use MIL-T-FDS*, special characteristics are unnecessary.

*Make It Like-The-F'ing Drawing Says

Trivial few/Vital many is properly applied to expense caused by errors, which doesn't necessarily have anything to do with special characteristics. I'll say it again: If the specifications reflect what's really needed, and you use MIL-T-FDS*, special characteristics are unnecessary.

*Make It Like-The-F'ing Drawing Says

And there is also MIL-TDS
Make It Like The Drawing Stupid:rolleyes:

And there is also MIL-TDS
Make It Like The Drawing Stupid:rolleyes:


It is easy for us to make these comments and have fun on this forum. I might even remember a couple of these.

But, we all know, there are a lot of variables, customer design problems, supplier issues, tooling issues, mating part issues....

In a pure world, we would always make it like the drawing, and that would always work. But, it doesn't always work like that...so we invented "special characteristics"...

It is easy for us to make these comments and have fun on this forum. I might even remember a couple of these.

But, we all know, there are a lot of variables, customer design problems, supplier issues, tooling issues, mating part issues....

In a pure world, we would always make it like the drawing, and that would always work. But, it doesn't always work like that...so we invented "special characteristics"...

Well Said!

Thank you all for a very informatve discussion. As I see it there is no right or wrong - black or white - answer to this issue. What is needed to control the output? That control can be designated as a "Special Characteristic". Thanks again for your input!:)

Thanks Jim, you made me realize something I have been missing. I should be correlating the Highest RPNs in my FMEAs to the critical (I refuse to use the word special in this post anymore), characteristics!

Just my two cents worth: Automotive stopped using the term "critical" many years back because the lawyers felt that it had product-safety connotations. The terms "significant" and "special" started to be in vogue to get around this.

Recall the adoption of nonconformance to replace defective for similar reasons. Lawyers interpret the word defective as an admission of guilt.

Just my two cents worth: Automotive stopped using the term "critical" many years back because the lawyers felt that it had product-safety connotations. The terms "significant" and "special" started to be in vogue to get around this.

Recall the adoption of nonconformance to replace defective for similar reasons. Lawyers interpret the word defective as an admission of guilt.

Yes plus "issues", "concerns", and "trouble reports"

I learned so much from reading this thread! Thanks to you all for your opinions, ideas and of course the humor. Thanks!:thanx:

I'm going to come down on the other side of this one, and say that the interpretation that there must be special characteristics is too invasive. My interpretation of "...shall identify special characteristics..." is that if there are any special characteristics, they must be properly identified as such, and I (or my customers) will decide whether there are any or not, not an outsider.

We are getting a surveillance audit this week. The auditor spent almost 30 minutes preaching to us that we "must identify a special characteristic even if the customer does not". We strongly disagreed. He called his managers who told him he was wrong. So, no finding. My belief is that if the standard intended us to contrive or invent a special characteristic, it would have made it clear by specifically stating it in the clause with something like this, "In the absence of a customer defined special characteristic, the organization shall define at least one and treat it as such."

We are getting a surveillance audit this week. The auditor spent almost 30 minutes preaching to us that we "must identify a special characteristic even if the customer does not". We strongly disagreed. He called his managers who told him he was wrong. So, no finding. My belief is that if the standard intended us to contrive or invent a special characteristic, it would have made it clear by specifically stating it in the clause with something like this, "In the absence of a customer defined special characteristic, the organization shall define at least one and treat it as such."

Makes sense to me.

We are getting a surveillance audit this week. The auditor spent almost 30 minutes preaching to us that we "must identify a special characteristic even if the customer does not". We strongly disagreed. He called his managers who told him he was wrong. So, no finding. My belief is that if the standard intended us to contrive or invent a special characteristic, it would have made it clear by specifically stating it in the clause with something like this, "In the absence of a customer defined special characteristic, the organization shall define at least one and treat it as such."

I agree with your viewpoint, but not the specifics of your post. I also agree with Jim Wynne's comment.

I agree, I do not believe the intent of the clause was to require a contrived SC on every product. Thatw ould add no value and just confuse things. I think the standard wants to make the point that many products have characteristics which are more significant for various reasons. It simply means that these significant characteristics should be identifed, whether the customer identifies them or not.

If they are significant, call them that. If they are not, don't. But we need to drop the philosophic positions that we won't if the customer doesn't. That misses the point.

PS: this was actually an expectation even in the QS-9000 era. It was discussed in the AIAG "blue books."

Maybe we can look at this problem from the other side…

Lack of SC could be a problem for the company.

Let’s imagine audit dialog:
Auditor: Does product have SC?
Auditee: No, it hasn’t.
Auditor: Your analysis didn’t show any SC, did it?
Auditee: Exactly.
Auditor: OK. What is important for your customer in this product?
Auditee1: Nothing. Customer didn’t designate anything as important characteristic.
Auditee2: I can’t say this in this way. Maybe hole diameter is important.
Auditee3: In my opinion length of the product is important. The length is important for most of our customers.
Auditee4: Looking on customers complaints – customers usually give us complaints because of the poor product hardness. So maybe it is important…

And here we have the beginning of the problem – it’s lack of priorities statement.
So – of course product needn’t to have SC, but every auditee has to know what is important and what is not important in the product. And they have to prove, every important characteristic is under control.

Maybe we can look at this problem from the other side…

Lack of SC could be a problem for the company.

Let’s imagine audit dialog:
Auditor: Does product have SC?
Auditee: No, it hasn’t.
Auditor: Your analysis didn’t show any SC, did it?
Auditee: Exactly.
Auditor: OK. What is important for your customer in this product?
Auditee1: Nothing. Customer didn’t designate anything as important characteristic.
Auditee2: I can’t say this in this way. Maybe hole diameter is important.
Auditee3: In my opinion length of the product is important. The length is important for most of our customers.
Auditee4: Looking on customers complaints – customers usually give us complaints because of the poor product hardness. So maybe it is important…

And here we have the beginning of the problem – it’s lack of priorities statement.
So – of course product needn’t to have SC, but every auditee has to know what is important and what is not important in the product. And they have to prove, everything important characteristic is under control.

No, I don't think that is right. The auditee does not have to "prove" anything, they have to show objective evidence. That slight language difference may seem like a very fine difference, particularly outside the USA, but it is important.

Auditors are supposed to evaluate evidence and render an assessment whether a process is effective. An auditor is not supposed to determine what the important characteristics are, he is to assess.

So, for example, let's say I evaluate manufacturing processes, customer complaints, scrap trends, and notice that a particular product characteristic causes a lot of problems. I interview operators and quality personnel, and the agree that feature causes a lot of problems. In fact, they show me a quality alert based on a recent customer complaint. I would definitely raise the question about why this is not identified as an SC - not because I think it should, but because the evidence indicates that it is.

I think this is the intent of the standard in this clause.

No, I don't think that is right. The auditee does not have to "prove" anything, they have to show objective evidence. That slight language difference may seem like a very fine difference, particularly outside the USA, but it is important.
Of course my english is not so good :o
For me: "Show me the evidence" and "Prove me" means almost the same.

Of course my english is not so good :o
For me: "Show me the evidence" and "Prove me" means almost the same.

Peters, your English is very good. But, because this is an important point, I wanted to explain it for those who were not sure.

For me it's difficult to imagine that some part doesn't have s.c. Any product is designed to serve some concrete purposes which are provided by complying the product with some main parameters.. geometric, electrical or smth else. The lowest nut has to correspond to (e.g.) inner diameter requirements which are considered to be the main parameter (s.c.) for this part and to be controlled during the production process.. (mainly for statistical process control). If somebody hasn't specified the s.c. for product it doesn't mean that they are don't exist... :confused: This is a theory, reality is not so simple one.. but.. I assume that there are logical reasons why you do not have s.c. for some product but "customer doesn't specify s.c." is not the reason.

For me it's difficult to imagine that some part doesn't have s.c. Any product is designed to serve some concrete purposes which are provided by complying the product with some main parameters.. geometric, electrical or smth else. The lowest nut has to correspond to (e.g.) inner diameter requirements which are considered to be the main parameter (s.c.) for this part and to be controlled during the production process.. (mainly for statistical process control). If somebody hasn't specified the s.c. for product it doesn't mean that they are don't exist... :confused: This is a theory, reality is not so simple one.. but.. I assume that there are logical reasons why you do not have s.c. for some product but "customer doesn't specify s.c." is not the reason.

There's a very simple reason that a job shop might not bestow the "SC" honorific: a belief that meeting all of the customer requirements will result in a product that satisfies the customer requirements. It may be true that there are certain characteristics the supplier might want to monitor more closely than others, but that's not always the case. Forcing a supplier into a corner with regard to analysis and reporting requirements is not a good way to do business. From the supplier side, claiming the absence of special characteristics just to avoid extra work, and failing to acknowledge the utility of statistical monitoring is also not good business.

I believe knowing your “special characteristics” should be part of every organization – required or not required by a certain standards… it is like your “key performance indicator” to continually improve or to maintain your status quo… and you can never do this, without knowing your special characteristics…

...and every organization has this special characteristic… it’s the DNA of every organization…

Hey all that sympathy with the auditor looks good in theory, try implementing in practice.
You should have sent your auditor to an English class, where they show the difference between "shall" and "must".
Its the preorgative of the design engineers to nominate even after using the special charecteristic chart.
You have taken it very well.

Hey all that sympathy with the auditor looks good in theory, try implementing in practice.
You should have sent your auditor to an English class, where they show the difference between "shall" and "must".
Its the preorgative of the design engineers to nominate even after using the special charecteristic chart.
You have taken it very well.
The only difference which relates to this is the difference between "shall" and "should"... “Shall” indicates a mandatory requirement.. “Should” indicates a recommendation..

The only difference which relates to this is the difference between "shall" and "should"... “Shall” indicates a mandatory requirement.. “Should” indicates a recommendation..

Correction: The more recent understanding is - Should indicates a requirement with significant flexibility in how it is implemented. Recommendations are generally found only in the ISO 9004:2000 standard (which is very worthwhile for everyone to read).

Does the language matter?
If it's in the standard, you either have to do it, or have a good reason why you don't.

Correction: The more recent understanding is - Should indicates a requirement with significant flexibility in how it is implemented.What Dvor transposed is the actual text of TS16949:2002. So, if "should" is supposed to be considered more than a simple recommendation, where is that expectation documented? I looked @ the Rules, FAQ's SI's, etc.... Could not find it. Is this something conveyed by "tribal knowledge"?:cool:

Does the language matter?
If it's in the standard, you either have to do it, or have a good reason why you don't.


The problem is that the auditor is not willing (may be insane) to make sense of a logical reason not to have a special charecteristic.
That is the reason we had to look for what it shall say, should say or must say....

What Dvor transposed is the actual text of TS16949:2002. So, if "should" is supposed to be considered more than a simple recommendation, where is that expectation documented? I looked @ the Rules, FAQ's SI's, etc.... Could not find it. Is this something conveyed by "tribal knowledge"?:cool:

Well, I will confess to a partial mea culpa...

Officially, the TS standard states:

In this Technical Specification, the word “shall” indicates a requirement. The word “should” indicates a recommendation.... Paragraphs marked “NOTE” are for guidance in understanding or clarifying the associated requirement.

I do know this has been communicated and emphasized frequently in some of the official trainings in Southfield (Detroit) by Plexus. However, that is not binding, even if they say that is the intent.

So, I reviewed the FAQ's, SI's and TS standard. The only thing I found was it was a former Ford CSR:

3.16 Should
Indicates a mandatory requirement with some flexibility allowed in compliance
methodology. Organizations choosing other approaches to satisfy a “should” must be
able to show that their approach meets the intent of ISO/TS 16949:2002.

However, the Ford CSR of Feb. 2008 modified that and followed the revised TS definition.

Sooo...:bonk:...my Bad. The definition ahs changed and I have to stay with the official documents, not the Plexus training emphasis...

PS: It would probably be prudent to not go overboard on theis official definition and simply dismiss "shoulds" as only 'recomendations." Proper consideration should still be given.