Machine learning improves remanufacturing

In the »Desire4Electronics« project, Fraunhofer IPA is working with 10 project partners to advance automated disassembly processes for the remanufacturing of small electrical appliances. The project runs from 2023 to 2025 and addresses United Nations Sustainable Development Goals.

People in Germany generate an average of 19.4 kg of electronic waste per person and year – and the trend is rising. Particularly in the case of small electrical appliances, many components containing valuable materials such as copper, polymers or lithium are still functional and could be remanufactured. Currently, however, the process is very time-consuming. The aim of »Desire4Electronics« is therefore to develop methods that automate and simplify the remanufacturing of components. Machine learning (ML) techniques, a subfield of artificial intelligence (AI), are to be developed and later applied to the entire remanufacturing process.

Interdisciplinary project consortium

The consortium is led by the Robotics and Assistance Systems department at Fraunhofer IPA. The Process Innovation project group at the institute's Bayreuth site is also collaborating on the research project. In addition, there are three other project partners: acp systems, Deprag Schulz and the »United Nations Institute for Training and Research«. In addition, six associated partners from the waste management industry support the project with their practical knowledge and can transfer important findings from the project directly into application.

Focus on sustainability

In the project, the research association is developing low-risk and intelligent automation solutions for remanufacturing. The focus here is on sustainability. Machine learning techniques will be used to recognize device types and joining techniques, and the latter will be checked using image and tool data. Based on the results, the project partners can develop multi-tools that can be used to solve various connection techniques for the disassembly of small electrical appliances. Methods for automated disassembly are also being explored. They can also establish the refurbishment process in industrialized countries and make it profitable, thus contributing to greater sustainability in production and consumption.