Skip navigation
  • +49 (0) 69 2556 1921
  • info(at)

Identify reusable materials with AI

The Recirce project uses artificial intelligence to identify plastics in complex products.

It is still difficult to separate plastics from a material mix for recycling. Established near-infrared spectrometry is reaching its limits for complex products with a high material diversity. This is where artificial intelligence (AI) is now to help develop a comprehensive recycling process. This is what the project "Digital Lifecycle Record for the Circular Economy" - Recirce for short - is all about. This includes a digital product passport that is intended to ensure transparency along the entire recycling chain. The combination of the product passport and a sensor-based sorting process supplemented by artificial intelligence is intended to ultimately channel more valuable resources into recycling. In concrete terms, a sorting system at the Fraunhofer IWKS facility, which uses infrared technology and 3D object design to analyze the composition of waste, will be supplemented by AI, which will be used to train the facilities machine learning using sensor data.

Project partners are the DFKI, the TU Darmstadt, the company Green Delta, the Fraunhofer IWKS and the company Papier-Mettler from Rhineland-Palatinate. The project is part of the DFKI Competence Center AI for Environment and Sustainability (DFKI4planet) and is funded by the German Federal Ministry of the Environment (BMU) as part of the "AI Lighthouses for Environment, Climate, Nature and Resources" funding program.


  • (19.10.2020)
  • Photo: © Fraunhofer IWKS