Repair and Yield Forecast for Electronics Productions Lines

OData support
Hajdu István
Department of Electronics Technology

The subjects of my thesis were the rework on Surface-Mount Technology (SMT) lines - focusing on the complex components (BGA); the resource planning and execution systems (ERP, MES) and the statistical estimate of yield.

The replacement of complex components is difficult. The companies may outsurce this process to an other party or scrap these failed materials. After the consideration of options for the reworking methods, the conclusion is, that it slightly affects the yield.

The ERP-systems are applications built of modules, which provide software solution for unified, real-time and integrated handling of production, logistics, sales, human resources management and controlling transactions. The MES is a combination of hardware and software components, which allows the management and optimalisation of production activities starting from the issue of orders to the production of end products.

In the following I got to know the methods for calculating the yield (FTY, RTY, DPU), and also a statistical estimation method, which provides a solution for estimating the yield at every quality control step in the process, and also the final ratio. We use the Poisson distribution derived from the Binomial distribution. It’s λ parameter equals the value of DPU (Defect Per Unit), and we define the random variable as zero so it results in an exponential dependence. I calculated the yield from one factory. Because of the large number and the low probability the statistical estimation of the yield approximates to this equation: Y=1-DPU.


Please sign in to download the files of this thesis.