DPMO and PPM
Defects per million opportunity (defects is the key word) and PPM- Defective parts per million (defective is the key word)
A process/product where a single unit of output has multiple opportunities for error/defect/failure(A) Vs a process/Product where a single unit of output has limited opportunities of error(B). for scenario A, DPMO is more suitable because it accounts for every defect that occurs in a unit, for eg. a data entry form(a unit of output) might have 50 fields to enter, each field could be considered as an opportunity for an operator to do mistake(defect) therefore accounting each defect becomes important. But why?? When we process millions of transactions, if we don’t track exactly where we are repeatedly going wrong within the 50 fields, it become extremely difficult to prioritize for us to work on it, Secondly, per example above, there is effort in entering every field accurately and so the DPMO measure brings out the actual score that determines the accuracy of work.
On the other hand, PPM measure defectives, in the example mentioned above, lets say we using PPM, the operator has entered 49 fields correctly and entered 1 wrongly, in such case it is considered as a defective form. What about 49 right fields? that’s the problem, in PPM its not accounted, it just says that the form is defective
Scenarios A : 100 units with 50 fields per form and we have say 50 defects(imagine there is 1 defect in each chart)
DPMO = (50/(100X50))X1000000=10000
99% accurate
Scenario B: 100 units with 50 fields per form and we have 50 Defectives (imagine there is 1 defect in each chart)
PPM = 50/100*1000000=500000
50% accurate
Therefore, I feel that DPMO over PPM will best suit where there are multiple opportunities to error in a unit of output