Digitalisation and Artificial Intelligence

Goal and Vision

The Digitalisation and Artificial Intelligence research programme is aimed at integrating automated experimentation, data infrastructures and advanced modelling into a unified, intelligent framework for materials discovery and manufacturing. In this way, materials can be designed, tested and optimised through closed-loop workflows that combine robotics, artificial intelligence and physics-based understanding, significantly accelerating the transition from concept to application. The programme’s focus is on data-driven materials engineering, i.e. establishing robust digital infrastructures for capturing and linking process parameters, material states, microstructure and performance (digital integration), enabling autonomous experimentation through self-driving laboratories (autonomous discovery), and developing predictive

models with quantified uncertainty to support reliable decision-making and inverse design (intelligent modelling). In addition, the programme will implement digital twins of manufacturing processes, allowing real-time monitoring, prediction and optimisation of material behaviour and performance in industrial environments. Experiments remain an essential component of the programme, providing high-quality data for model training, calibration and validation, while also enabling continuous feedback between physical and virtual environments. The research activities are supported by advanced robotic platforms, modular automated laboratories and high-performance computing infrastructures, combining CPUs and GPUs to enable large-scale data processing, simulation and AI model development.

Main research lines

  • Development of standardised, interoperable, and machine-readable data pipelines for additive manufacturing, enabling traceable integration of process parameters, material states, microstructure, and performance across experiments, simulations, and industrial workflows.
  • Creation of autonomous experimental platforms combining robotics, in-line characterisation, and closed-loop optimisation to accelerate the discovery, formulation, and processing of polymer-based materials.
  • Development of trustworthy reduced-order and AI-based models that combine predictive accuracy with rigorous uncertainty estimation, enabling robust decision-making, inverse design, and accelerated exploration of complex materials and manufacturing spaces.
  • Implementation of digital twins that integrate real-time data, physics-based models, and artificial intelligence to monitor, predict, and optimise manufacturing processes, improving quality, adaptability, and resource efficiency.

Research groups