If you are a Fabless start-up and are ramping, you will struggle without some sort of analysis capability for your data in production. You can decide to purchase a system that manages all the data, or alternatively, manage the data yourself and hire an engineer whose job it is to generate reports from the data. He or she may use Excel or some other desk top software tool. Whichever they choose, they will have a full time job just providing the right information from the data coming in, and it will get harder and harder for one person as you ramp, so you may have to hire an extra person also.

shutterstock_125411825

Or you could go for a system solution for about a third of the price of an engineer. The engineer you hire will now spend only a few minutes a day generating reports and the rest of the time adding real insights and value.

This is what a good cloud-based Yield Management system should be able to do for you as you ramp:

  1. Send reports designed by you to your Inbox every morning, tracking your yield, bin distributions and the performance of key tests
  2. Alert you on anything going amiss in production (you tell it the rules), any time of the day
  3. Allow you to get to any amount of detail to discover the root cause of a yield issue
  4. Take care of all your data management and allow you to scale without being concerned about the data system slowing down or creaking (the beauty of the cloud)
  5. Allow you to quickly correlate issues across manufacturing (e.g. from final test back to fab) so you can help your foundry with information that will improve your yield
  6. Enable you to analyse across hundreds of wafers’ worth of data and compare performance by week, by month or by quarter
  7. Handle any data format you have in production: STDF, WAT, csv, .tst etc
  8. Allow you to be in control of quality through providing algorithms like GDBN and PAT to your wafer maps (even handset makers are requiring automotive-like quality now)
  9. Provide sophisticated analysis tools for test and product engineers, for example sensitivity analysis and virtual retest to simulate changing test limits
  10. Allow you to manage your lots on hold and generate your statistical bin limits.

Cloud based yield systems are revolutionising Yield Management and are very affordable now for fabless start-ups. Hire engineers to add value in the lab and not to spend most of their time looking for information in reams of new and old data.

Contact MFG Vision today to see a demo of such a system.