Profile of Customer
This UK company is customer of yieldHUB’s for the past five years.
- Hundreds of packaged lots per quarter tested in two subcons in Asia
- Each lot tested multiple times across several testers
- Each lot has 1000’s of units
- Certain bins only are rescreened
- Yield or analysis from datalogs not accurate unless the files are combined into one version for each lot
- This was a manual process and subject to huge engineering hours per week and manual error.
So could yieldHUB generate an accurate picture for each lot by automatically combining the datalogs correctly, taking into account which hardware bins were actually retested?
Each datalog that we receive is processed in conjunction with all other datalogs for the same lot. yieldHUB detects the Run Type (eg first test or rescreen or QA) and takes into account the rescreen algorithm. For example only bin 7 and bin 8 may be re-screened.
yieldHUB engineers also monitor the process twice daily in case there are any issues with the data. Sometimes files arrive late or datalogs are marked incorrectly. In the past there were bugs in the consolidation scripts due to not taking into account extreme cases. These are always fixed with customer advice and close collaboration and review.
In Q3 2018 we are providing the customer with manual editing to update run types and lot IDs where mistakes are made by the subcon. This will greatly reduce maintenance requirements and will include audit capability that records who has updated the data using the editing tool.
“We are able to collaborate in real time with our team in Shanghai, enabling us to release new products much faster than before”
Can yieldHUB Help You?
Contact yieldHUB today! Our global sales and support team will be happy to help.