Auditing Inventory Levels - Computer Count vs. Physical Inventory

[noboost4you]

Let's forget for a minute the argument about whether this is a valid NC.

Let's instead take this as an opportunity to evaluate the effectiveness of your system.

How accurate does the information in your ERP system need to be?

To answer that question, you need to know how people use the information. What kinds of plans and commitments do they make based on the information? What do they do with it?

Do any of those plans and commitments affect the customer? (I bet they do.)

If so, what is the potential for inaccurate information to adversely affect those plans and commitments? If the potential is there, you probably ought to find ways to improve the accuracy of the information.

How do you deal with/prevent the scenario where the company rep promises 32 widgets overnighted when there are only 27 on the shelf? If this scenario can occur or has occurred, then you should look for ways to improve the accuracy of the information.

Have there ever been customer complaints about failure to meet promised ship dates? Can those complaints be traced back to promises made on the basis of inaccurate information? If so, fix the information.

How satisfied are your customers with your level of service? And how do you know? If you are actively seeking customer feedback and they think your level of service is good, you may not have a problem. If they rate your level of service low, you may be seeing the effects of inaccurate information.

Have promised ship dates ever been missed because the promises were made on the basis of inaccurate information? If so, you should fix the information.

Do you have a documented procedure for maintaining inventory counts? If so, does it say anything about accuracy of information? If so, you must meet the requirements of the procedure. If you don't, change the practice or the procedure.

Is there a policy or an expectation about the accuracy of inventory information? (It would be good if there were.) If so, does the current situation meet that policy or expectation? If not, changes need to be made.

If you can answer all these questions and see no negative effects of inaccurate information, then you've done a reasonably thorough analysis and you can respond to the NC by saying that your current system for maintaining accurate inventory information is effective and no corrective action is needed.

But if you find negative effects in running through this question list, then you have a problem.

While I appreciate the time you took to write all of this information, I am not here to discuss the rammifications of having an ineffective ERP system. I want to know if this is a valid nonconformity based on the ISO 9001:2000 Standard that does not state the physical count must match the computer count.

Yes, many people use the ERP system everyday to check inventory levels for customers. Have there ever been problems? You bet. Have we lost customers because they were unsatisfied? Probably. However, every company goes through this whether they are ISO registered or not.

If a customer orders 32 widgets and we promise overnight delivery, but only have 27 widgets in stock, we will catch that immediately. When we do catch this problem, production will notify Customer Service and they will call the customer. We will get to the customer before the customer contacts us after a few days without a package. It's at this point where an Inventory Adjustment form will be filled out and the ERP system corrected.

 

[Doug Tropf]

Re: Auditing Inventory Levels

Here is a snipet of 7.5.5 from our quality manual:

Components and finished product are referenced by manufacturer part numbers. Inventories of all product and components are maintained in the [Company] ERP System to determine purchase points and ensure that product is available for distribution."[/I]


I believe your auditor is justified in issuing a minor. Your quality manual, as quoted above, states that your ERP system should be capable providing information that can be used to determine purchase points and ensure that product is available for distribution. The auditor found verifiable evidence that the inventory levels indicated in your ERP system are not accurate. As others have pointed out, this situation has the potential to create a host of problems.

 

[Evert Taube]

Interesting discussion !!
All audit findings shall be based on objective evidence so 2 out 4 with wrong numbers is not just "bad luck". An increased sample-size (~10) would give more information of the situation !!

Secondly does it feel safe to not rely on your computer figures? As you said nobody´s perfect but the intention is to have good control over what you have in stock to minimize delivery problems.
Inventory turnover is often an important KPI (key performance indicator) and is of course based on forecast in sales, lead-times from suppliers and own manufacturing.

Regards from an "control-freak"

 

[noboost4you]

Re: Auditing Inventory Levels

I believe your auditor is justified in issuing a minor. Your quality manual, as quoted above, states that your ERP system should be capable providing information that can be used to determine purchase points and ensure that product is available for distribution. The auditor found verifiable evidence that the inventory levels indicated in your ERP system are not accurate. As others have pointed out, this situation has the potential to create a host of problems.

Yes, the ERP system is capable of providing information. Whether that information is correct is debatable.

So by our quality manual we backed ourselves into a corner, is what you're saying? If we removed that one sentence from the manual, can any future auditor gripe about the difference in inventory counts? I still don't see how the Standard can dictate physical counts must match computer counts. What about companies that don't have an ERP system? What if the only thing they rely on are what's on the shelf?

Based on our customer satisfaction surveys in which 20% were returned completed, we were ranked 4.5 out of 5. So if this were a serious issue, we wouldn't have that kind of satisfaction.

 

[CliffK]

I know this isn't the answer you seek. Sorry if I'm causing you frustration.

I think there's a better way of evaluating this NC than the "no basis in the standard" argument. Down that road lies an endless argument, which nobody wins. I think I'm proposing a lot more persuasive method for denying the NC if it is, in fact, invalid. Read on.

While I appreciate the time you took to write all of this information,

You're welcome.

I am not here to discuss the rammifications of having an ineffective ERP system.

It's not about the system. We could be having this discussion if you kept inventory data on 3X5 cards.

I want to know if this is a valid nonconformity based on the ISO 9001:2000 Standard that does not state the physical count must match the computer count.

Yes, many people use the ERP system everyday to check inventory levels for customers. Have there ever been problems? You bet. Have we lost customers because they were unsatisfied? Probably.

So the NC in question is an opportunity to improve, right? It's up to you to decide if it's the right thing to do to seize the opportunity. What would it cost to make the ERP data more accurate? What would the payback be? If the payback is greater than the cost, why would you not do it?

However, every company goes through this whether they are ISO registered or not.

I'm not sure about the "every company" part. But then if one accepts that it's inevitable then one has created a self-fulfilling prophecy.
What about continual improvement?
What about the oft-quoted statistic that it costs 10 times as much to get a new customer as it does to retain an existing customer?

If a customer orders 32 widgets and we promise overnight delivery, but only have 27 widgets in stock, we will catch that immediately. When we do catch this problem, production will notify Customer Service and they will call the customer. We will get to the customer before the customer contacts us after a few days without a package. It's at this point where an Inventory Adjustment form will be filled out and the ERP system corrected.

So you have a method for containing the effects of bad inventory data in the ERP system.

Did you explain that to the auditor?

The real question before you is if your customers are satisfied with this method of doing business (assuming you truly know--do you?). If they are, then your method of containment is adequate and the NC is not valid. If they're not, then you have a problem that needs fixing.

A secondary question is about process measurements. Do your measurements show that your order fulfillment process is meeting its objectives? If so, the NC doesn't mean much today. If you raise the bar on the fulfillment process, it might come into play tomorrow, though. An argument based on objectives isn't as strong as one based on customer sat, but it should fly.

As a poster in another forum observed, it's really about whether the NC adds value, not what subclause you hang it on.

Oops. I missed the point about your ERP system and the quality manual. If the QA manual says something about the ERP system, then you have probably created a situation that I call "audit bait," or sometimes, "gotcha bait." You should probably revise the QA manual, 'cause there's no need to go into this level of detail and lots of reasons not to.

 

[noboost4you]

I know this isn't the answer you seek. Sorry if I'm causing you frustration.

I'm just frustrated in general; not because of you. You have made us think in greater depth about the overall system.

Did you explain that to the auditor?

Unfortunately no. It's one of those things that "you wish you would have said" at the time, but it didn't come to you at that point.

The real question before you is if your customers are satisfied with this method of doing business (assuming you truly know--do you?). If they are, then your method of containment is adequate and the NC is not valid. If they're not, then you have a problem that needs fixing.

Of the 500 surveys sent out, we received 20% back and once tallied, we were rated 4.5 out of 5. We must be doing something right.

A secondary question is about process measurements. Do your measurements show that your order fulfillment process is meeting its objectives? If so, the NC doesn't mean much today. If you raise the bar on the fulfillment process, it might come into play tomorrow, though. An argument based on objectives isn't as strong as one based on customer sat, but it should fly.

This is actually one of our quality objectives. We grade ourselves on how often we ship complete line items of an order. In 2006 it was 39%. We finished 2007 at 48.9%. While we aren't where we'd like to be, we are continually improving that number. It's an on-going process that we strive for on a daily basis.

As a poster in another forum observed, it's really about whether the NC adds value, not what subclause you hang it on.

Well if it doesn't hang on anything, how can it be a nonconformance? It could simply be an observation saying "Hey, you might want to look into this" but not throw out a nonconformance because he felt the need to do so.

Oops. I missed the point about your ERP system and the quality manual. If the QA manual says something about the ERP system, then you have probably created a situation that I call "audit bait," or sometimes, "gotcha bait." You should probably revise the QA manual, 'cause there's no need to go into this level of detail and lots of reasons not to.

I am looking into industry standards for inventory accuracy percentages. Once I find a correct industry standard, we will revise the quality manual to state such.

 

[CliffK]

I'm just frustrated in general; not because of you. You have made us think in greater depth about the overall system.

I sense that. But I think we can see our way out of the thicket.

Unfortunately no. It's one of those things that "you wish you would have said" at the time, but it didn't come to you at that point.

Gosh, that's never happened to me. (I'm lyin')

Of the 500 surveys sent out, we received 20% back

That's high.

and once tallied, we were rated 4.5 out of 5. We must be doing something right.

Here we've reached the nub, the very crux of the exercise. If this rating is acceptable to you, then there's objective evidence that you are meeting your business goals/objectives. At this point I would argue that the NC is meaningless, except for the QA manual issue, discussed below.

This is actually one of our quality objectives. We grade ourselves on how often we ship complete line items of an order. In 2006 it was 39%. We finished 2007 at 48.9%. While we aren't where we'd like to be, we are continually improving that number. It's an on-going process that we strive for on a daily basis.

More evidence that the quality system is working. You have some sound reasons here for arguing that your method for containing bad ERP data is effective.

Well if it doesn't hang on anything, how can it be a nonconformance? It could simply be an observation saying "Hey, you might want to look into this" but not throw out a nonconformance because he felt the need to do so.

Well, it can hang on your quality manual, too. Or on your procedures and work instructions or any other document that describes the work method or required results.

That said, a proper NC needs to contain a description of, or clear reference to, the requirement that's being violated along with a clear description of the condition that is violating the requirement. If that's not in the NC, you have a valid ground for complaint. Your auditor should be able to give you this information. If he can't or won't, get another guy to do it next time.

Last point on this: while I'm feeling too lazy to undertake the exercise, I bet I could find a subclause to hang this on. We might argue over it, but then it just gets into an interpretation issue. Yuck.

Real last point on this: that's not to say you shouldn't ask to see the "shall" if in doubt about the validity of an NC. But the shall should be visible in the text of the NC.

I am looking into industry standards for inventory accuracy percentages. Once I find a correct industry standard, we will revise the quality manual to state such.

I generally counsel people not to go there in the QA manual. The QA manual should contain the three mandated topics along with enough information to give your successor a good start if you win the lottery this Friday and decide to retire on Monday. Anything more is a waste of keystrokes and asking for trouble. (OK, I admit that's a bit strong, but it's an approach that's worked for those I've taught it to.)

If you want to use your manual as a sales tool, create a one pager (we loves our customerses. Oh yes we does) and put the guts into a companion document.

 

[Doug Tropf]

Re: Auditing Inventory Levels

"I still don't see how the Standard can dictate physical counts must match computer counts. What about companies that don't have an ERP system? What if the only thing they rely on are what's on the shelf?"

In most instances, ISO and other standards do not provide details on how an organization controls it's processes, but rather requires that such controls be effective. If your organization believes it's inventory system is effective then that should be your response to the audit finding.

I've generally approached reasonable audit findings as having the potential to make our company better. There does appear, from what you've presented in this thread, to be some opportunity for improvement with your inventory system.

 

[SteelMaiden]

I just have to throw this in...
Multiple comments have been made about high satisfaction marks on a customer survey, 4.5 out of 5. Just remember, you get answers that match the questions you ask, and the data recieved may not have any reflection on the problem at hand, or it may only reflect the perception of the customer, not cold hard facts.. I had a customer rate us at 100 out of 100. I know for a fact that we have been late a couple of times. Why did he give us perfect marks in his own supplier scorecard? Because we are less late than all his other suppliers, therefore he felt we should get full credit. OK, good for us, but not very objective or helpful.

 

[noboost4you]

I just have to throw this in...
Multiple comments have been made about high satisfaction marks on a customer survey, 4.5 out of 5. Just remember, you get answers that match the questions you ask, and the data recieved may not have any reflection on the problem at hand, or it may only reflect the perception of the customer, not cold hard facts.. I had a customer rate us at 100 out of 100. I know for a fact that we have been late a couple of times. Why did he give us perfect marks in his own supplier scorecard? Because we are less late than all his other suppliers, therefore he felt we should get full credit. OK, good for us, but not very objective or helpful.

Ah yes, the topic of an effective customer satisfaction survey is another area of concern, but just not yet.