Supplier Defects Per Million Metric - Best way to depict monthly supplier performance

I am trying to figure out what could be the best way to depict monthly supplier performance.


I would appreciate any suggestions on how to measure and portray supplier quality.

At incoming inspection lets say we receive a lot with 500 parts. We sample 10 parts.

1) If all 10 are good we allow all 500 parts to go to the assembly line.

2) If even 1 out of the 10 is bad, we reject the entire lot of 500 and consider 500 pieces as defective in our DMR (damage material report) database.

If the 500 parts are allowed to go through to the line (based on the incoming/receiving inspection), if the assembler/operator finds 1 or 2 parts defective, these 1 or 2 pieces are added as defects to the DMR database.

Now at the end of the month when we want to portray the month's supplier quality, we pull up information from the DMR database which gives us the defects per million (DPM).

Parts defective divided by the parts received and the result multiplied by a million. The problem with this metric is that there is a large variation in the data since it is a combination of 500 pieces rejected and 1 or 2 line defect pieces rejected and in some cases the DMR'd pieces are as many as 10000 and give an unnecessarily high number of DPM (defects per million).

Now we do have a seperate metric for individual suppliers. What I am talking about above is one DPM number to reflect supplier performance as a whole.

Thanks in advance!

AB,

You're quite right that under the system as you describe, there's a lot of
noise in the data. (This system is what several of my clients' customers use,
so I have had the opportunity to analyze & try to explain it!)

Several options that might help.

1. If the line operator finds defects, put the balance of the lot into "suspect"
or defective category. This makes the data more consistent. but not more
accurate. (Some customers do this, too!)

2. Have the suspect lots sorted (regardless of where the defect was found,
incoming or at line. Then get the actual data resulting from the sort, and
use that for defects per million. Analytically better, depends on trusting the
supplier to sort. Some suppliers will abuse this and "find no more defects". Sooner or later you will figure out which ones are abusing the system and take
some other action to fix them.

If you really want to know, sort the suspect lots yourself and charge back
to the supplier. This also gets you material quickly. Some suppliers may
decline the charge. Perhaps you give them the option to come on site
and sort at your location where you can keep an eye on the process.
In the spirit of good parts, 100% of the time, this may be warranted.
In some industries, though, the suppliers may tell you to get lost.

I am willing to be harsh and give the supplier some financial pain for sending
nonconforming product. Some of our automotive customers have learned
that this gets suppliers' attention quickly!

Regards,
Brad

Brad

Thank you very much for taking the time to reply to my message. I really appreciate it.

Unfortunately, my problem is bigger than it appears. On the bright side, our contracts with all suppliers are such that even if we find 1 part bad out of a lot of 5000, we have the authority to reject the entire lot and send it back to the supplier and they dont complain. Then they might sort it at their end and resend parts.
My problem is that we dont have the resources or time available to warrant a 100% sort. So that is not an option given our present setup.

Anyways, thanks again.

AB

Supplier defects per million metric

Ab,

Going in we let our suppliers know that if we find defects on the line, they have two choices -- to come in and sort the material within a specific time period, or to let us bring in a 3rd party to sort and change back to the supplier. We also change back for lost productivity.

We do very little incoming inspection and it was very painful at first, both for us and our suppliers; fortunately we were in ramp-up. Things have smoothed out considerably since then.

This process gives us a fairly accurate estimate of supplier dppm because someone is sorting 100%.

Ben