Associate Professor University of Florida GAINSEVILLE, FL, United States
The advent of 3D advanced packaging has introduced significant challenges in physical inspection and reliability analysis. As the industry progresses towards more complex and miniaturized structures, ensuring the integrity of these packages becomes increasingly critical. Traditional inspection methods often fall short in addressing these complexities, leading to a heightened interest in non-destructive testing (NDT) techniques. Methods such as thermal and 3D X-ray imaging have been employed; however, they face limitations in resolution and the ability to detect defects in densely packed components. To overcome these challenges, integrating Artificial Intelligence (AI) and machine learning (ML) into inspection processes has become pivotal. These technologies enhance the capabilities of NDT methods by enabling automated defect detection and analysis, thereby improving accuracy and efficiency. Moreover, the implementation of multimodal analysis and data fusion—combining data from various NDT techniques—provides a more comprehensive understanding of potential defects and material properties. This approach enhances the robustness of monitoring performance and addresses the intricate needs of modern packaging inspection. This presentation will delve into the state-of-the-art inspection methods, exploring the associated challenges and outlining a research roadmap for this emerging field.
