Calculating PPM for Customers - Is this fudging PPM #'s?

[cleverfox]

Hi again everyone. Lets see what you all think of this.

In calculating PPM for customers, how do the rest of you handle this?

My experience has been to use only qty of returned parts against qty shipped.

But what if your customer sorts and scraps 'x' amaount of your product? Shouldn't this be counted against against your quality rating also?

My little shoulder devil and angel are battling it out, although I know better! But I wanted to see how the rest of you weigh in. I know how management has defined it in the various companies that I have worked for.

cleverfox

 

[howste]

I think you already know the answer to this - defective parts per million should include all defective parts...

 

[bpritts]

cleverfox,

You refer to parts 'returned' vs parts that are defective. We might get
a full container of parts returned if the customer finds a few that are
defective; if so, we would sort and charge only the defective parts,
not the total return.

Meanwhile I agree w. howste-- if the customer has scrapped out some
bad ones you should charge them, too, whether returned or not.

Brad

 

[Claes Gefvenberg]

Howard & Brad are absolutley correct imo... There is one thing that complicates matters though, and Brad already touched it:

How easy is it to get information about all defective parts...? from all customers...? all the time...? Do you always get those figures from the customer? What if the customer sorts the gadgets you provide and find them to be usable, maybe with some difficulty, even if they are not acc. to spec.... and then neglects to tell you about it?

What I'm saying is that the figures will be of little value if the basic data varies. You will not be able to trust any trends.


/Claes

 

[gpainter]

Do not fudge the numbers! This is one thing that data analysis is about, bring accurate data to the table, so problems and trends can be identified and appropriate actions taken. DO NOT SUGAR COAT DATA! If you have to do this, your companies system is not at the maturity level it needs to be. It happens in many companies, you are constantly walking on glass.

 

[D.Scott]

cleverfox,

You refer to parts 'returned' vs parts that are defective. We might get
a full container of parts returned if the customer finds a few that are
defective; if so, we would sort and charge only the defective parts,
not the total return.

Brad

This happens to us as well however some of our customers track the entire container as rejected and technically unusable which counts against their tracking of our PPM. Like Brad says we might only find a few "bad parts" and that is what we consider as part of our PPM but what do you do when the customer rejects a bin of 200000 parts and counts them all as PPM on your rating?

Once again, we are dealing with customer perception. If the customer rates your PPM at a higher rate than you, you are going to be stuck with the higher PPM. At least in the eyes of that customer.

Dave

 

[Sam]

Hi again everyone. Lets see what you all think of this.

In calculating PPM for customers, how do the rest of you handle this?

My experience has been to use only qty of returned parts against qty shipped.

But what if your customer sorts and scraps 'x' amaount of your product? Shouldn't this be counted against against your quality rating also?

My little shoulder devil and angel are battling it out, although I know better! But I wanted to see how the rest of you weigh in. I know how management has defined it in the various companies that I have worked for.

cleverfox

Why do you let your customer "sort & scrap"? You have no control over the process, so why bother to count the scrap. Counting inaccurate results is no different tnen not counting at all.

 

[ben sortin]

The only reason to track PPM is to achieve zero defects. Please consider the following:

< 200 PPM requires flawless containment of suspect product.
< 100 PPM requires timely corrective and preventive actions.
< 60 PPM requires requires validation of effective advanced quality planning.
< 30 PPM requires the customer to understand that product and process improvements take time and money.

Zero defects requires the supplier and the customer to master "soft quality."

Given all that, never fudge the numbers. Teach your customer(s) to manage the breakthroughs and your quality people to manage the plateaus.

 

[db]

I worked with a company that machined cast parts. They had carts with two scrap bins. One was for "foundry" scrap, the other was for "manchined" scrap. They tracked what parts were not good from the foundry, and what parts they ruined.

When looking at customers sorting and scrapping, make sure you are not considering the scrap that the customer created.

...Counting inaccurate results is no different then not counting at all.

Garbage in = garbage out. If you can't trust the numbers, you can't trust the numbers.

 

[bpritts]

Several excellent comments have been made about the "why" of tracking
ppm that are worth summarizing; there are several reasons to do this.

1. As an internal performance measure PPM can give you a common index
of your performance. You'd like to think that it's accurate in an engineering
precision sense, but it probably isn't. As Claes and Sam suggest, you can't
really trust the customer's figures on defectives. Some of the defects
will *never* be discovered.

The better you get, the more the measured PPM will represent "rare events",
where normal statistics don't work very well anyway. Lots of auto suppliers
have multiple months of zero reported PPM, then get little (or sometimes big)
spikes; using averages, etc., can be misleading. That said, having a best
efforts PPM number will help you keep score, so it's worth doing.

One of my clients used to track "months since a customer complaint", just
as you've seen people do with accidents. That had the benefit of simplicity
and helped communicate to the people.

2. Customer reported PPM's, as Dave points out, are vitally important even
if statistically meaningless. If that's the number the customer rates you by,
it's data of a different sort. If your customer thinks you're a 20000 PPM company
because you had a bin of 20,000 bolts rejected for 10 pieces of mixed stock,
you may have a problem, especially if they're comparing you to people selling
them transmissions that are 10 PPM. (and they're not sophisticated enough
to realize that a $500 transmission competes in a different playing field
from a $.02 bolt. ) Get your sales guy or gal going!

3. Lastly, gpainter's point is well taken. Report the facts accurately. If some
"spinning" needs to be done, let the sales department do it.

Best regards,

Brad