Solving tough semiconductor manufacturing problems using data mining

Raising product yield and quality, or quickly solving problems in a complex manufacturing process is becoming increasingly more difficult. Process control, statistical analysis, and design of experiments have established a solid base for a well tuned manufacturing process. However, the dynamic "next-tier" problems such as multi-factor and nonlinear interactions, intermittent problems, dynamically changing processes or installing new processes, multiple products and the sheer volume of data can all make quickly finding and resolving problems an overwhelming task. Data mining technology applied to data analysis can increase product yield and quality to the next higher level by quickly finding and solving these problems. Case studies of semiconductor wafer manufacturing problems are presented. A combination of self-organizing neural networks and rule induction is used to identify the critical poor yield factors from normally collected wafer manufacturing data. Subsequent controlled experiments and process changes confirmed the solutions. Wafer yield problems were solved 10/spl times/ faster than standard approaches; yield increases ranged from 3% to 15%; endangered customer product deliveries were saved. This approach is flexible, easy to use, and can be appropriate for a number of complex manufacturing processes.