EUREKA EURIPIDES2 Failure Analysis 4.0

Important results from the project

The goal of the project was to develop a comprehensive mostly automated process chain for failure diagnostics integrating self-learning data analysis and defect detection and localization, standardized material analysis and centralized data acquisition and correlation to electrical data. The main contributions of the Swedish consortium were the development of an X-ray source with a resolution well beyond state-of-the-art delivering superior quality 3D images of components from partner semiconductor suppliers and automated void detection in X-ray images using ML metods.

Expected long term effects

Reduced time spent on failure analysis was achieved by the Machine Learning methods of defect analysis developed. More efficient use of existing equipment was achieved by introducing a universal sample holder and automated data transfer. Improved accuracy in failure analysis has been achieved by e.g. higher resolution equipment, automated sample preparation, and the ML methods developed for failure analysis which also enables more samples to be analyzed. Thanks to these improvements faster product development and innovation cycles are expected to be achieved.

Approach and implementation

Sweden was integrated in the project after the project was approved in other countries and the integration worked very well since our competences complemented the existing ones by adding X-ray to the techniques investigated. The Swedish consortium activities had its focus on defect identification in WP3, Task 4.5 “Ultra-high resolution X-ray imaging” (Task leader RISE) of WP4 (co-lead by RISE), combination of the developed tools and methods in WP5 (co-lead by Ericsson), and summary of the project achievements in WP6 (lead by Ericsson). The project was extended by 6 months (COVID).

External links

RISE projektsidor svenska RISE projektsidor engelska Projektets publika sidor