Valeria Borodin
IMT Atlantique
Actions
Associate Professor at IMT Atlantique, whose research focuses on the development of methodological and algorithmic approaches to support decision-making in complex industrial systems
Links
Hybrid Multiscale Plasma Simulation and Machine Learning for Bosch Etching
This talk introduces a hybrid modeling framework for the Bosch etching process that combines multiscale plasma-surface simulations with machine-learning surrogates. Physics-based models generate datasets linking process parameters to key CFₓ species and passivation behavior, while ML models enable fast prediction and analysis of etch outcomes. The proposed approach represents a step toward physics-aware process optimization and data-driven control without relying on purely empirical tuning.
From reference process flows to executed routes: Predicting slowdowns in an R&D cleanroom
In semiconductor manufacturing R&D environments, product routes are initially sketched using a reference process flow, while their effective realization unfolds progressively during execution through local decisions, lot filiations, and device-team holds, which can be assimilated to slowdown and hold-induced interruptions. In this paper, we propose a predictive modeling approach to anticipate such slowdown regions along R&D routes. This approach corresponds to a static view, in which at the beginning of lot processing, it estimates slowdown risks over the whole route horizon.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top