From: Mark Maral
Date: 17 Apr 2000
Time: 11:33:43
Stan,
Thank you for the quick response. Also, sorry for the duplicate posts. I wasn't sure if the first one took. Just to preface my response, we have 3 different OEMs providing repaired drives to a third party company that does a receiving inspection for us. Currently, they are inspecting 100% of the drives they receive.
In answer to your questions:
1. Monthly total is around 2700 drives per month. this breakdown is Supplier A=350, Supplier B=325 Supplier C=2100 drives. Daily totals are for Supplier A=20, Supplier B= 20, Supplier C= 100. Each supplier supports, on average, 10 different part numbers that can be received on any given day as part of the daily total.
Another note is that the current monthly quality is for Supplier A=95%, Supplier B= 99.5%, and Supplier C= 99.2%
2. Not quite sure how to answer this. Ultimately, the goal I would have would be to do away with this third party Receiving Inspection and rely on the OEM's repair process and their own internal self-surveillance and implement if needed a sampling plan at the OEMs repair facility. Your thoughts on this point is would be appreciated.
3. Sampling would be on a daily basis. Currently, as the drives are being inspected if a reject is found, it is noted, but inspection continues and good units are accepted. There are no lot rejections based on a failure. We have never in fact considered lots or making each days receipt a lot.
5. Currently, I do not believe that there is a significant quality problem between various part numbers. We have attempted to qualify the OEM's repair process and for the most part each repair line is a fairly standardized process.
I believe there are a number of sampling plans that work well in the mfg. process where everything for the most part are in lots. In sampling repaired product I am having a hard time with this. Since we are in a service organization providing repair product to support field defectives, and management wants to keep inventory to a minimum, they are very much against establishing a lot size and then rejecting a whole lot of drives due to the accepted sampling criteria. Thus a I think I am working against the tide, but feel there is a better way to sample repaired product without compromising the pipeline.
Thank you for your help.
Mark Maral