Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

When to notify the customer of a possible nonconforming condition. 4

Status
Not open for further replies.

QAFitz

Materials
Jul 21, 2005
121
0
0
We make fasteners and related hardware to ASTM material Specs. and ANSI and Customer provided drawings. If the item is a repeat then we make enough to stock and ship as needed. Occasionally there are a few issues that arise and I'm looking for comments and/or suggestions. Keep in mind that many customers have no problem spreading the noncompliance word but rarely tell the whole story (sounds like the 4th estate doesn't it?). Keep in mind that the original production run was inspected and tested per the applicable material Spec and accepted, and, some of the fastener material Specs establish that one (1) strength test can qualify up to 800-pcs.

Two of the recurring issues are:
- When we pull items from inventory and gage them some won't accept the Go thread gage or, will accept the NoGo thread gage. The original lot was accepted based upon an AQL.
- When we pull items from inventory we usually check the hardness as a precautionary measure. Sometimes we find a one (1) to two (2) point higher or lower than allowed hardness. The original lot was tested per the material Spec (AQL) and accepted.

Comments and suggestions are greatly appreciated.
 
Replies continue below

Recommended for you

Not sure of your question. Based on your comment below
Occasionally there are a few issues that arise and I'm looking for comments and/or suggestions.

Yes, things happen when our QC department receives items in the company I work for that have been stated by the vendor as being compliant and were not compliant. The important take away from this is we will work with the vendor to correct or disposition the nonconformance, and we move on. If the pattern is repeated by the vendor, we may visit the vendor to see what the problem is and attempt to correct it at their facility. Otherwise, I shop for a new vendor.
 
metengr: These are items that we inspected and accepted once, shipped to a customer, and put the remainder in stock to be used at a later date. At the later date, inspections and hardness are checked and sometimes we find an issue.
 
What this shows is that your process is not sufficiently controlled to justify the sampling rate that you are using.

You need to investigate the manufacturing processes to find out why they result in non-conforming parts.

As for when to notify the customer. You notify them as soon as you identify the problem.
 
I agree completely with everything MintJulep wrote.

You need to verify if your parts have variation, or maybe your testing equipment, or your operAtors use slightly different methods that contribute to variation.
 
QAFitz,

I also concur with the previous comments. It's obvious that you are trying to be conscientious about this, so I recommend that you take the advice and use it to improve your company's processes. Variation is probably the key word, and the goal should be understanding whether it is from the manufacturing process, or the inspection process, or both.
 
I think "identify the problem" is a bit of a grey area.

I would not formally inform the client until I've verified that there is a problem and accurately identified the nonconforming condition. If it turns that your "problem" was isolated or traced back to improper testing during your spot checks, etc, you've created doubt and probable cause for the customer. At this point, he's probably going to a) make you go back and do supplemental testing on the supplied parts at your expense, or b) hammer you in court for nonconforming products that resulted in lost time, wasted materials, etc.

Just make certain that you don't create problems that you aren't sure exist yet.
 
Sounds like you need a tighter original AQL level. You have more variation in the product that was assumed when the test plan was set up.
There are two separate areas to engage your customer with;
1. desired level of testing vs cost vs passing second inspection
This is a quality management issue
2. sources of high variation in original product
This is one that you should be working on internally already.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube
 
Ideally we test to make sure that we do have a problem. Once we have a problem we develop a couple of solutions to offer the customer. Then we approach the customer, explain the problem we have and offer the solutions we've developed.

For decades we were caught between our suppliers and our customers and quality issues. What finally worked for us was getting together with a major customer for us and for the industry and writing a nice, complete set of specifications. We very carefully spelled out all the defects we had ever seen, we develop methods to measure and otherwise quantify them and we publicized it throughout the industry.

This has solved a great deal of our problems. However, we did have a very significant advantage in that we were dealing with a huge number of relatively low-cost parts on an ongoing basis.


Thomas J. Walz
Carbide Processors, Inc.

Good engineering starts with a Grainger Catalog.
 
Status
Not open for further replies.
Back
Top