Journal of the Microelectronics and Packaging Society

Advanced packaging technologies are rapidly evolving to meet the semiconductor industry’s increasing demands for higher performance, miniaturization, and lower power consumption. Among these technologies, wafer-level packaging (WLP) has emerged as a key solution due to its superior capability in achieving compactness and enhanced performance. However, predicting the reliability life of WLP remains a significant challenge due to its complex structure and various environmental factors. Traditional reliability life prediction methods, such as physicsbased modeling and accelerated life testing, are limited by high costs and long time requirements. To address these limitations, artificial intelligence (AI), particularly machine learning (ML) algorithms, have gained significant attention. This study discusses recent trends in ML algorithms for reliability life prediction in advanced packaging, focusing on unsupervised learning, supervised learning, and hybrid learning approaches. Additionally, the paper provides insights into potential future research directions.

Keywords Machine learning algorithms, Reliability life prediction, Supervised learning, Unsupervised learning, Hybrid learning, Advanced packaging