SOLVING SMALL SAMPLE RECIPE GENERATION PROBLEM WITH HYBRIDWKRCF-PSO

The cost of the experimental setup during the assembly process development of a chipset, particularly the under-fill process, can often result in insufficient data samples. In INTEL Malaysia, for example, the historical chipset data from an under-fill process consist of only a few samples. As a result, existing machine learning algorithms for predictive modeling cannot be applied in this setting. Despite this challenge, the use of data driven decisions remains critical for further optimization of this engineering process. In the proposed framework, the original weighted kernel regression with correlation factor (WKRCF) is strengthened by normalizing the input parameters and employing the Particle Swarm Optimization (PSO) as weight estimator. It is found that PSO gives flexibility in defining the objective function as compared to the iteration technique of WKRCF. Thus, an assumption on noise contamination to the available training samples can be implemented. Even though only four samples are used during the training stage of the conducted experiment, the proposed approach is able to provide better prediction within the engineer’s requirements as compared with WKRCF. Thus, the proposed approach is beneficial for recipe generation in an assembly process development.