Infinineon Semiconductor AI Lab
Joint Interdisciplinary “Infineon Semiconductor AI Lab” with joint projects in the areas of AMS-Design, ML/AI and Multiphysics Modeling of Integrated Circuits
About
The partnership between the Technical University of Cluj-Napoca (UTCN) and Infineon Technologies Romania (IFRO) began in 2013 with the goal of strengthening and expanding competencies in the semiconductor field. This successful collaboration brings together academic expertise and industry leadership to drive innovation. The partnership has resulted in numerous joint research projects focusing on semiconductor chip design, artificial intelligence, and microelectronic device simulation. Beyond technology, the cooperation has sparked significant educational initiatives. This includes the creation of specialized microelectronics study programs tailored for integrated circuits and systems within the automotive industry. Infineon and UTCN actively encourage students to specialize in this essential, high-growth field by providing internships and scholarships, helping them reach their full potential in microelectronics. A pivotal milestone in this collaboration is the establishment of the Infineon Laboratory within the Artificial Intelligence Research Institute (AIRi@UTCN). This facility serves as a cutting-edge hub where advanced AI research meets hardware excellence, allowing students and researchers to develop hardware-aware AI solutions and next-generation semiconductor applications.
Common Projects UTCN-INFINEON
Title: "Automation of Differentiated Sample Lot Analysis Based on Artificial Intelligence Methods"
**Period: ** August 1, 2024 – July 31, 2025 **Description: ** The analysis of differentiated sample lots (split lot analysis) is one of the key tasks in technology development. It consists of evaluating intentional modifications made to the semiconductor chip manufacturing process to determine the impact of various settings on the output parameters of the resulting devices and to decide how to adjust the technological process to improve overall performance.
In this project, we aim to automate specific parts of this analysis. Previous evaluations have shown that AI and Machine Learning techniques can be highly effective for:
- Predicting distributions for split lots that have not been implemented.
- Automated discovery of distribution anomalies.
- Calculating the process window for PCM (Process Control Monitor) parameters.