SPC for Eddy Current Testing

[a.temel]

Hi,i would like to ask your opinions/advices on some matter,

We are producing cold drawn steel bars. And one of the steps of the production is crack control operation (Eddy current). During the 3. party audit (IATF 16949), auditor told that we should be doing process control using data gathered from crack control operation too(minor non-conformity). Problem is, ratio of material seperated during crack control operation is highly dependant on the raw material (%5 to %50 percent seperated). Since we buy our steel from companies that we cannot change due to many reasons, data gathered from the process cannot be used to evaluate suppliers and since its an attribute data set, it cannot be used to measure process capability (As far as i know). What im getting at ,in this case there is no reason to do SPC because there is no point. I guess the question is, what we are going to do when when we have a non stable process output due to raw material and we cannot choose / change or improve our suppliers.

Sorry for the long post.This my first one.Thanks in advance.

 

[Bev D]

Did the auditor state the specific requirement in 16949? (I am nor well versed in this standard). Is there a specific requirement to have control charts on everything or on specific types of processes?

You are correct that using SPC for your eddy current test is not appropriate for all of the reasons you just stated.
There are some situations where having this data might be helpful overall:

- supplier quality monitoring: this will tell you which suppliers have better control over their raw material - but it isn't very actionable if you are truly constrained in your supply base.​

- procurement planning: knowing what the actual yield variation is might help your procurement group do a better job at procuring the necessary quantities.​

​

Of course this isn't process control in the strict sense of the phrase and intent. but SPC isn't always the best choice.

sounds like you need to have a discussion with your auditor...

 

[a.temel]

@Bev D

First of all thanks for your reply. Auditor told us this non conformance verbally. He wanted us to correct it and said he will check it at next years audit.

As for our suppliers, we are truly constrained,and procurement was never a problem. Since we also sell our raw material directly, we always have enough raw material.
Also any material separated during crack control operation can be sold other customer who doesn't require crack controlled products.

 

[Bev D]

A ‘verbal’ is not a nonconformance. My cynical 2 cents: Your auditor is trying to pull a fast one to impose his uninformed agenda...

 

[Steve Prevette]

I am always wary when someone says "there is no point" to doing SPC. I'd at least look to is there a business case here. Are you having problems with cracks? Are they causing rejects or rework? In such a case you may want to understand if you have a stable process or not. I'd agree though it is only "an oral" in the audit and then the only reason to deal with the data at all is - what is the business case.

 

[Bev D]

There is a point there but my understanding of the OP's question is that the auditor is 'asking' for SPC on the Eddy current testing results. this would in effect monitor the quality of the incoming material which is going to be non-homogenous and with a constrained supply base there may be very little in practice that the OP can do about this incoming variation other than 'screen out the bad material and order enough to ensure that they don't have an availability problem due to excessive yield loss. I don't think it's a matter of can SPC be applied, but m ore a matter of whether or not it help drive control or improvement. SPC is not the only effective control mechanism.

There may be an opportunity to understand the effectiveness of the eddy current testing in catching the bad material: are there cracks in the product at downstream operations? this may be appropriate for SPC if there is some level of level of cracking. However, if there is cracking downstream it may be more effective to perform a MSA on the eddy current testing to improve it's detection rate and/or the material culling process.

 

[a.temel]

A ‘verbal’ is not a nonconformance. My cynical 2 cents: Your auditor is trying to pull a fast one to impose his uninformed agenda...

My suspicion is that,during the audit he might not be informed correctly since i'm the only one who knows the basics of SPC in the company and i didnt participate in audit.

There is a point there but my understanding of the OP's question is that the auditor is 'asking' for SPC on the Eddy current testing results. this would in effect monitor the quality of the incoming material which is going to be non-homogenous and with a constrained supply base there may be very little in practice that the OP can do about this incoming variation other than 'screen out the bad material and order enough to ensure that they don't have an availability problem due to excessive yield loss. I don't think it's a matter of can SPC be applied, but m ore a matter of whether or not it help drive control or improvement. SPC is not the only effective control mechanism.

There may be an opportunity to understand the effectiveness of the eddy current testing in catching the bad material: are there cracks in the product at downstream operations? this may be appropriate for SPC if there is some level of level of cracking. However, if there is cracking downstream it may be more effective to perform a MSA on the eddy current testing to improve it's detection rate and/or the material culling process.

We had no complaints related to this matter. But also we and our customers has no way of verifiying the result of the crack control operation (Visual control is not enough to see the cracks we are testing for). Also eddy current machine by design doesnt return any numeric value, machine only seperates the bars according to sample bar we identified before the serial production. So i dont know if there is any way to perform an MSA study.

 

[Mike S.]

Perhaps I don't understand the OP's process -- in fact, I am pretty sure I don't understand the OP's process very well at all, so take what I offer within that context.

Having said that, is there no value to looking at the statistics, process capability, etc. of % of cracked material? That is not an attribute that is variable data.

 

[a.temel]

Perhaps I don't understand the OP's process -- in fact, I am pretty sure I don't understand the OP's process very well at all, so take what I offer within that context.

Having said that, is there no value to looking at the statistics, process capability, etc. of % of cracked material? That is not an attribute that is variable data.

We are cold drawing hot rolled steels through a die, then straighten it, then check for cracks at the end of the line. And cracks are almost all of the time related to the raw material and we cant change our suppliers. So there is no way achieving an improvement using data gathered from crack control step.

 

[Bev D]

@a.temel : so for clarity, you are using eddy current testing after you form the metal? if the eddy current tester indicates a crack the part is scrapped? and the reason a control chart (in your opinion) is not useful is that the material is from a constrained set of suppliers and (in your opinion) you have no leverage to improve their material quality? Also you are accepting that the eddy current testing is catching most - if not all - of the cracked material as your Customers have not complained about any cracked material escaping your facility to their facility or their customers - correct? (just as an aside, there is a way to perform an MSA on the eddy current testing, you are probably simply not aware of it). And can you confirm that your auditor asked specifically for a control chart or were they asking for other forms of statistical analysis (MSA, Capability, Yield trending, etc.) sometimes people conflate control charts and other forms of statistical analysis into one big lump of SPC....

@Mike S. : % of cracked material is attribute data and not variable data.