A genetic algorithm based knowledge discovery system for the design of fluid dispensing processes for electronic packaging

In the semiconductor manufacturing industry, fluid dispensing is a very common process used for die-bonding and microchip encapsulation in electronics packaging. Understanding the process behaviour is important as it aids in determining appropriate settings of the process parameters for a high-yield, low cost and robust operation. In this paper, a genetic algorithm (GA) based knowledge discovery system is proposed to discover knowledge about the fluid dispensing process. This knowledge is expressed in the form of rules derived from experimental data sets. As a result, appropriate parameters can be set which will be more effective with respect to the required quality of encapsulation. Rules generated by the GA based knowledge discovery system have been validated using a computational system for process optimization of fluid dispensing. The results indicate that the rules generated are useful and promising in aiding optimization of the fluid dispensing process in terms of better optimization results and shorter computational time.