CENTRALE LYON - Doctorat Modélisation prédictive de matériaux polymères recyclés par inférence statistique et apprentissage machine hybride.
Core
Developing predictive models and hybrid statistical/statistical inference frameworks to ensure stable production quality of recycled polymer materials despite raw material variability.
Role type
PhD researcher (predictive modeling & machine learning)
Builds
Robust predictive models (digital twins) for real-time process control in polymer transformation lines.
Domain
Materials science / Polymer engineering / Industrial process control
Deliverable
production ML models
Required skills
Machine learning, Statistical inference, Physics-based modeling, Python programming, Rheology knowledge
Preferred skills
Digital twin development, Experimental data fusion, Simulation tools
Responsibilities
Design original learning structures fusing experimental and synthetic data, Develop predictive models for real-time process piloting, Predict final material properties (viscosity, granule behavior) based on machine settings.
Seniority
PhD candidate (36 months)