Abstract
The FA2IR project will investigate important applications of Artificial Intelligence (AI) methods to databases in microelectronic failure analysis (FA). The target is to get these FA databases AI-ready and to develop improved FA4.0-AI-based methods, e.g. for image and measurement data analysis, text classification, etc.
Although AI is the subject of various FA-related publications, the approach of this project regarding the database landscape is unique due to the implementation of the FAIR-data principle (Findable, Accessible, Interoperable, Reusable).
This work will include deep-dive data evaluation, e.g. finding and labeling failures in images as well as finding analysis reports describing them. The novel methods will result in shorter development times for new products, a faster and exacter reaction to field failures, as well as improved collaboration between partners of the value chain by an open approach.