View Full Version : How do I answer an audit finding I don’t agree with?
Hodgepodge 9th October 2008, 02:37 PM A three-day second-party audit generated a minor finding. AS9100 section 8.2.4 requires sampling plans to be “statistically valid” and “shall preclude the acceptance of lots with known conformities”. Our final inspection guidelines detail the use of MIL-STD-105E and also states “If nonconformance is discovered, that drawing characteristic shall be inspected 100% on all parts in the order.
The finding was that an invalid sampling plan is used. The auditor explained that the plan must be a C=0 plan.
My root response was “Documentation error. Incorrect wording in process guidelines. The intent of the process is zero acceptance, however, lot acceptance measures are not delineated as such.” We were using the sampling tables in the Mil-Std to come up with a quantity to use as a sampling quantity and the intent was to reject lots with known nonconformities. We have since purchased and implemented an ASQ Press sampling plan. This response was rejected. By document error, I mean the document was written incorrectly.
Let us not take into consideration that I’m sure you can use MIL-STD-105 with an acceptance number of zero. I suppose the next “why” would be, “Why was the process written incorrectly?” The answer would simply be, “The guy that wrote it doesn’t know what he is talking about.” Why is an unqualified person determining process parameters?
My question is: Does it really need to go to this level? I was learning when I wrote the process, and I’m still learning now. I know that I am acting defensive, but, hey, I'm human. Thanks in advance for any guidance in this area.
Coury Ferguson 9th October 2008, 02:44 PM The Auditor is correct. You need to have a sampling plan that is C=0.
No way around this. It doesn't mean you can't use MIL-STD-105E for the sampling, but you can't use it for Acceptance or Rejection. I was cited for the same thing during the Stage 2 Assessment.
Sidney Vianna 9th October 2008, 02:46 PM I will only mention that CB's are expected (and being continually scrutinized) to ensure that the certified organizations have effective corrective action processes, which implies meaningful root cause analysis.
From the description of the situation you offer, it seems that the auditor is doing what is expected of him/her.
Shallow root cause (oxymoron?) is a plague that infects and affects the world of quality management. While it won't be eradicated, it needs to be combated, if we are to improve our systems.
Caster 10th October 2008, 12:28 AM And just for fun...does a C=0 sampling plan actaully "...preclude the acceptance of lots with known conformities” ?
bobdoering 10th October 2008, 07:24 AM My question is: Does it really need to go to this level? I was learning when I wrote the process, and I’m still learning now. I know that I am acting defensive, but, hey, I'm human. Thanks in advance for any guidance in this area.
I can not say why an auditor would reject the response that you misinterpreted the requirements, and with the corrected interpretation you have implemented the correct sampling plans. Yes, they may jump back to issues of training and qualification (and, one might ask is there opportunity for improvement there - such as more training). It is hard when you are the one that is getting hit with a non-conformance not to get a bit defensive. But the point of the audit is not to be punitive, but to get the system in order. I think you have every right to ask the auditor when they reject a response as to the specific issue that they feel has been inadequately addressed. They can not tell you how to address it, but they may suggest the type of objective evidence that will show that it - and its root cause - have been addressed.
Coury Ferguson 10th October 2008, 07:31 AM Let us not take into consideration that I’m sure you can use MIL-STD-105 with an acceptance number of zero. I suppose the next “why” would be, “Why was the process written incorrectly?” The answer would simply be, “The guy that wrote it doesn’t know what he is talking about.” Why is an unqualified person determining process parameters?
My question is: Does it really need to go to this level? I was learning when I wrote the process, and I’m still learning now. I know that I am acting defensive, but, hey, I'm human. Thanks in advance for any guidance in this area.
The answer is yes you should go that far into the Why's.
What you have answered in the Corrective Action is just a symptom, not Root Cause. Just my opinion here.
Randy 10th October 2008, 08:00 AM My root response was “Documentation error. Incorrect wording in process guidelines. The intent of the process is zero acceptance, however, lot acceptance measures are not delineated as such.” We were using the sampling tables in the Mil-Std to come up with a quantity to use as a sampling quantity and the intent was to reject lots with known nonconformities. We have since purchased and implemented an ASQ Press sampling plan. This response was rejected. By document error, I mean the document was written incorrectly.
Let us not take into consideration that I’m sure you can use MIL-STD-105 with an acceptance number of zero. I suppose the next “why” would be, “Why was the process written incorrectly?” The answer would simply be, “The guy that wrote it doesn’t know what he is talking about.” Why is an unqualified person determining process parameters?
My question is: Does it really need to go to this level? I was learning when I wrote the process, and I’m still learning now. I know that I am acting defensive, but, hey, I'm human. Thanks in advance for any guidance in this area.
Never go to a person, always go to a process. saying that "this person did this" or "this person failed..." shows weakness in the cause analysis process. Determining cause is exactly like water going downhill, it will always seek the path of least resistance...and people are always that path.
Not knowing everything, but basing "my" reasoning upon your statements I'd say we had a lack of identifying required compentencies as a "cause"...something had to be done, someone had to do it , and they didn't know what or how. That's basically what you said. So it took knowledge and the ability to demonstrate the use of that knowledge and it didn't happen.
Always look to "process".
amanbhai 10th October 2008, 09:11 AM Being an MR of testing, I have been audited by one the accrediation body sometime back.
I have the previlage to receive an CAR agianst 4.7.1 Note 1 (b) which is used for guidelines.
The external assessor has the concern that we as a third party lab does not address the ISO 17025:2005 guidelines which says" 4.7.1 Note 1 b) Preparing, packaging, and dispatch of test and/ or calibration items needed by the customer for verification purposes"
I had a long debate with that auditor for raising an CAR not against the standard but rather against guideline of the standard?
Now, I just need to know what other quality pros opinion about that?
How do I response to the audit finding I don't agree with?:thanks:
Coury Ferguson 10th October 2008, 09:15 AM Being an MR of testing, I have been audited by one the accrediation body sometime back.
I have the previlage to receive an CAR agianst 4.7.1 Note 1 (b) which is used for guidelines.
The external assessor has the concern that we as a third party lab does not address the ISO 17025:2005 guidelines which says" 4.7.1 Note 1 b) Preparing, packaging, and dispatch of test and/ or calibration items needed by the customer for verification purposes"
I had a long debate with that auditor for raising an CAR not against the standard but rather against guideline of the standard?
Now, I just need to know what other quality pros opinion about that?
How do I response to the audit finding I don't agree with?:thanks:
I am familiar with 17025, but not an expert. As far as I understand, there can not be a NC generated on a Note. The note is just clarification of a requirement, not the requirement, in my opinion.
AndyN 10th October 2008, 09:21 AM And just for fun...does a C=0 sampling plan actaully "...preclude the acceptance of lots with known conformities” ?
Well, dependent on the AQL that has been set.......the tighter the level, the greater the number of samples taken in a lot and the rules say if you find one defective, you reject the lot........
But it's all in the selection of the AQL, based on an understanding of producers risk (capability) compared to users risk, right?
Without that, it's a bit dicey, I believe.........:mg:
Hodgepodge 10th October 2008, 01:03 PM Never go to a person, always go to a process. saying that "this person did this" or "this person failed..." shows weakness in the cause analysis process. Determining cause is exactly like water going downhill, it will always seek the path of least resistance...and people are always that path.
Not knowing everything, but basing "my" reasoning upon your statements I'd say we had a lack of identifying required compentencies as a "cause"...something had to be done, someone had to do it , and they didn't know what or how. That's basically what you said. So it took knowledge and the ability to demonstrate the use of that knowledge and it didn't happen.
Always look to "process".
Thanks Randy,
So changing the process guidelines requiring the use of a C=0 plan is merely the containment action. Continuing down the cause chain we find the competency to perform the requirements of the position to not be fulfilled. I have been learning on the job with no formal training. To have management pay for training materials or formal training is like trying to get milk from a bull. My corrective action on this is going to have to be something I can actually do. I am learning here and now, with all of your help. The corrective action doesn't have to be immediate, but there is a requirement for corrective action projected completion date. I suppose I will run a distant date up the flagpole and see if anyone salutes.
Thanks to everyone!
Dean Frederickson 10th October 2008, 01:18 PM Thanks Randy,
So changing the process guidelines requiring the use of a C=0 plan is merely the containment action. Continuing down the cause chain we find the competency to perform the requirements of the position to not be fulfilled. I have been learning on the job with no formal training. To have management pay for training materials or formal training is like trying to get milk from a bull. My corrective action on this is going to have to be something I can actually do. I am learning here and now, with all of your help. The corrective action doesn't have to be immediate, but there is a requirement for corrective action projected completion date. I suppose I will run a distant date up the flagpole and see if anyone salutes.
Thanks to everyone!
I would pay to see that:lmao::lmao::lmao:
Randy 10th October 2008, 10:01 PM Thanks Randy,
So changing the process guidelines requiring the use of a C=0 plan is merely the containment action. Continuing down the cause chain we find the competency to perform the requirements of the position to not be fulfilled. I have been learning on the job with no formal training. To have management pay for training materials or formal training is like trying to get milk from a bull. My corrective action on this is going to have to be something I can actually do. I am learning here and now, with all of your help. The corrective action doesn't have to be immediate, but there is a requirement for corrective action projected completion date. I suppose I will run a distant date up the flagpole and see if anyone salutes.
Thanks to everyone!
Competency does not = Training
Competency has to do with the demonstration of knowledge and skill. Knowledge can be gained from education, taining or experience and skill is gained from experience. You can acquire what you need to do the job in multiple areas.
Now if your management doesn't wish to provide training then the next NC should be "lack of committment as evidenced by lack of appropriate resource allocation" and "lack of a competence process" as evidenced by your inability.
See? Everything becomes a system and not a personal issue.
On top of this we could probably also go into "ineffective corrective action process" as another NC.
It's amazing how things can compound and develop their own synergy.
Of course the auditor would be called a horses "A" even though it was the organization and it's top management that shot itself in the foot.
The use of the system can stop "egg sucking" really fast.
Big Jim 11th October 2008, 12:15 AM A slightly different perspective to the original question "How do I answer an audit finding I don't agree with". I'm making these comments without judging on the merits of the findings. In other words, I'm not saying the auditor was right or wrong.
As the other responders have pointed out, you need to come to terms with the finding, and discussing it here is a good way to expand your knowledge and thinking. You should always start with discussing it with the auditor. Often he can provide greater insight as to why he feels it is a problem. Also, sometimes you can point out additional information that may make a difference.
If, after you have fully studied it, you still don't agree, you can appeal it. I understand that all CB's are required to have an appeal process. You do have recourse if you truly feel that the auditor is wrong. You should be able to contact your CB and ask how to go about it. Sometimes it can be worked out over the phone without going through the formal appeal process.
That said though, sometimes it is easier to just roll with the punches than it is to get worked up about an appeal.
bobdoering 11th October 2008, 10:08 AM Randy has good points. When you are responsible (or feel that you are) for an NC, and you are hanging out their without resources or other management commitment, it is painful. It is hard not to internalize that pain. It is clearly management responsibility to provide the resources for you to do the job. In many ways, the auditor should have discerned that problem and written it up during the audit. That would have helped your situation a great deal. Right now, management thinks you are the owner of this NC. No matter how you answer it, it is your problem. However, if management commitment was written up, the could not skate on that. They would be the owners, and they would have to show how they are going to provide the resources. In fact, if I knew this NC was coming during an audit, I would have encouraged the auditor to write that up. Better coming from them....
Sometimes management will not listen to the people right in front of them, but will hang on every word of some guy who just walked in off the street.:cool:
Geoff Withnell 14th October 2008, 11:37 AM And just for fun...does a C=0 sampling plan actaully "...preclude the acceptance of lots with known conformities” ?
If c=greater than zero, then it is possible to accept a lot that is known to have had at least c nonconformities or non-conforming parts. If c=0, then by the definition of c=0, we have not found any nonconforming parts in the sample, and have no KNOWN nonconformities. Unknown, is unknown.
Geoff Withnell
Caster 14th October 2008, 09:16 PM .......Now if your management doesn't wish to provide training then the next NC should be "lack of commitment as evidenced by lack of appropriate resource allocation" and "lack of a competence process" as evidenced by your inability........See? Everything becomes a system and not a personal issue
I assume you are speaking as an external auditor. You could get away with writing an NC against management for "no support", but it sounds like if the poster tried that during an internal audit at his company, it would be a career limiting move.
Caster 14th October 2008, 09:28 PM If c=greater than zero, then it is possible to accept a lot that is known to have had at least c nonconformities or non-conforming parts. If c=0, then by the definition of c=0, we have not found any nonconforming parts in the sample, and have no KNOWN nonconformities. Unknown, is unknown. Geoff Withnell
Thanks Geoff, you are technically correct of course, but I would not expect much luck with this line of reasoning after I shut down an assembly line with a bad "unknown" part.
I have found very few people have any idea that there is risk with any sampling plan, never mind two types of risk.
Most seem to think that there is some magic sampling plan that will let any untrained person casually glance at the top 2 or 3 parts in any lot size and guarantee the rest are all OK.
|
|