Informational Nonconformances and Defects - Operator Error, System Error, or both?

[bobdoering]

The way I look at it is I do not blame the operator if the defect was caused by something that was not considered in the system. But, if the system specified that the operator is responsible for a task, the training has been done, and the operator does not perform that task, then you have operator error.

Now, as far as corrective action, you may find the risk or frequency of that error may justify a change in the system to further limit operator dependency. But a100% operator independent process just means machines will cause the problem or the people responsible for determining and performing the PM will be the next problem. A 100% operator independent process also means less employment (even lights out plants a few folks hanging around). But, that is an argument for the social responsibility standards people to address.

 

[Matt Swartwood]

I typically practice the Apollo approach to root cause analysis. I want to see due diligence to address the system where it all possible and operator error is the very last resort. We have our corrective actions set up where the only way you can use operator error as the root cause is if all system controls are in place to prevent the defect and the operator blatantly disregards procedure or bypasses current failsafes.

I'm not saying our way is the right way or the only way, but we have been very successful with it so far...

 

[Bev D]

But a100% operator independent process just means machines will cause the problem or the people responsible for determining and performing the PM will be the next problem.

I'm not sure of your point.

are you saying that any error proofing we do won't reduce the defect rate because the cause will just shift from a human to the error proofing device to break down? The only way we don't reduce the defect rate is if the error proofing device is just as prone to 'break down' as the operator was.

Or are you saying that any reduction in defect rate isn't worth the effort because we can never achieve zero?

I always pursue reduction in defects regardless of the fact that I will never achieve perfection - I've gotten nicely close to it at times. (and NO, I don't go after a .0001% trivial effect defect rate when I have a 5% defect rate that is very upsetting the customer. occuring - I use common business sense to prioritize the work.

 

[bobdoering]

I'm not sure of your point.

Let me clarify. Operators - even when well trained - are capable of making mistakes - whether accidental, whether changing the process on their own thinking they are "helping', via fatigue (either from work duties or what they did outside of work) or simple carelessness. They are responsible for their work.

So, to eliminate these root causes, you would have to eliminate the operator. That is possible to do in some scenarios. It may be more costly than using the operator (to the point of being unprofitable), but possible nonetheless.

If may require errorproofing - such as poke-a-yoked automated systems. But, even as the staunchest proponents of poke-a-yokes have found, you need to constantly verify them, because they wear out. So, human error has been replaced with mechanical error, to be supervised by a unfailing PM system. Until such time, though, you can take credit for incredibly low failure rates - and employment (but who is counting that?)

Are you saying that any error proofing we do won't reduce the defect rate because the cause will just shift from a human to the error proofing device to break down?

I am saying it will not eliminate defects for that reason. Will it reduce it? Perhaps - perhaps not.

The only way we don't reduce the defect rate is if the error proofing device is just as prone to 'break down' as the operator was.

It is not the only way. One other way is that the error proofing device 'breaks down' because people expect it to work, and it starts to fail. It can make defects for quite a long time. More than a human? Maybe.

Or are you saying that any reduction in defect rate isn't worth the effort because we can never achieve zero?

No I am not. Have it 24/7.

I always pursue reduction in defects regardless of the fact that I will never achieve perfection - I've gotten nicely close to it at times. (and NO, I don't go after a .0001% trivial effect defect rate when I have a 5% defect rate that is very upsetting the customer. occurring - I use common business sense to prioritize the work.

It is a great goal. And if you can eliminate humans in the process - and their inherent human error - and it makes common business sense, then I am sure it will happen. However, it can be divergent economically.

 

[bobbelden]

What if an employee who is packaging parts knowingly raids the MRB cage and ships 5 bad parts to the Customer? The customer has since found the parts and is requesting corrective action. What would be the cause and corrective action?

 

[Stijloor]

What if an employee who is packaging parts knowingly raids the MRB cage and ships 5 bad parts to the Customer? The customer has since found the parts and is requesting corrective action. What would be the cause and corrective action?

Without having any knowledge and details about the context in which this error occurred, I would say that rarely people "willfully" do these things.

My first concern would be: "did someone pressure this person to complete a shipment to get stuff out the door?"

Difficult to answer without all the specific details.

Stijloor.

 

[bobdoering]

What if an employee who is packaging parts knowingly raids the MRB cage and ships 5 bad parts to the Customer? The customer has since found the parts and is requesting corrective action. What would be the cause and corrective action?

That's the kind of thing that makes customers demand locked MRB cabinets of everybody...even when it is not an issue for some (like us). It is maddening...

What is the cause? Somewhere there is a perceived pressure (true or not) that making shipment is more important than anything else. Customers can create such pressures - in spite of their proclaimed desire for good product. JIT will really paint you in a corner. Not sure what this specific case is, though.

 

[Mike S.]

What if an employee who is packaging parts knowingly raids the MRB cage and ships 5 bad parts to the Customer? The customer has since found the parts and is requesting corrective action. What would be the cause and corrective action?

In that case, I don’t know enough to say.

But I have had an AS9100 certified firm, a subsidiary of a HUGE corporation, admit verbally (after we caught bad parts at our receiving dock) to knowingly shipping me rejected parts because their company’s highest officer told them to do it. On the official corrective action report it was reported as a “mistake” resulting from a “desire to help” support our company. I have a different word for it.

Don’t think it doesn’t happen – probably more than you know. If it looks like a duck, and quacks like a duck…open yourself up to the possibility that you just might have a duck on your hands, even if you don’t want to believe it.

 

[Bev D]

and this is why we must have all of the facts. In this case, dropping this supplier if at all possible is worth consideration. YOU are ultimately responsible for your supplier's quality...who knows what this guy is having people do. Depending on your ultimate Customer for this part you might consider bringing them in if you want to simple 'fix' the supplier. AS9100 Customers are pretty intolerant of this.

 

[Jim Wynne]

and this is why we must have all of the facts. In this case, dropping this supplier if at all possible is worth consideration. YOU are ultimately responsible for your supplier's quality...who knows what this guy is having people do. Depending on your ultimate Customer for this part you might consider bringing them in if you want to simple 'fix' the supplier. AS9100 Customers are pretty intolerant of this.

The question comes from the supplier's point of view. As others have suggested, the motivation for taking known bad parts is important to know. It seems to me that this was either an obsessive-compulsive person who couldn't stand to see a partial box, or someone who had a gun at his head.