Yield and DPMO - what can they tell us
The experts decided to use DPMO to show why these boards behave that way in production. The facts of the products and quality were:
When engineers calculated DPMO figures for termination (T), placement (P) and component (C) defects, very interesting results came out.
Engineers were surprised; this was something they hadn't expected. It came clear that line 2 quality capabilities are much higher.
Line 1 is producing almost six times more termination defects compared to Line 2. Also other defect categories have same trend.
The results were reported to managers.
Common understanding was that actually first step to take is to find out why Line 2 is so much better and the learning should be utilized in Line 1. If Line 1 could produce as good quality as Line 2, yield for product A would be 97% instead of current 92%.
What should be underlined is that DPMO cannot replace Yield everywhere.
From yield figures we can see immediately that product B requires more resources for rework and re-testing.
Remarkable costs savings could be achieved in long run when also line 2 processes are improved.
DPMO metrics can be used in various levels and it is very efficent when analyzing where are the real quality problems in those two lines
(material, certain component packages, equipment, etc.).
Previous page <-