Cloud-Native Digital Twin Architectures
Supervisors: Gianluca Bontempi, Gian Marco Paldino
Digital Twins (DTs) are virtual replicas of physical systems that are becoming essential in Industry 4.0 for monitoring, simulation, and optimization. Machine Learning Operations (MLOps) bridges the gap between laboratory ML models and robust, scalable, and maintainable models required for live DT environments. To achieve the scalability and real-time responsiveness required by modern DTs, MLOps is best implemented on a cloud platform.
The research topic could be related to developing a cloud-based digital twin architecture for real-time monitoring, forecasting, and optimization of physical systems. This architecture, once completed, would integrate physical simulators with real-world data streams, collect sensor data from IoT devices, processing large volumes of time-series data, and perform machine learning pipelines on cloud platforms, implementing closed-loop control systems for autonomous decision-making.
Therefore, required knowledge include python programming, core ML libraries like pandas, numpy, scikit-learn, basic data preprocessing, model training, evaluation metrics understanding and git fundamentals. We expect you to quickly learn the basics of Docker, Databases (eg. TimescaleDB), Cloud Deployment (eg. AWS), MLops (eg. MLFlow) and Streaming (eg. Kafka).
Tools & Resources: