Utilize resources
New technologies for the circular economy
AI and the digital product passport play a central role in the circular economy. Without smart algorithms, sustainable and therefore cost-saving production is hardly conceivable. But there is one hurdle: the data.
The efficient use of resources is an important lever for sustainable business today. SMEs in particular face the challenge of reducing material and energy consumption while remaining competitive. In order for the green transformation of the economy to succeed, the Federal Environment Minister had to get SMEs on board, as they make up 99.5% of companies. This is where the Green-AI Hub Mittelstand initiative comes in.
The program is aimed at SMEs and aims to help them use AI to make more efficient use of resources. David Möller manages the company 4Packaging. The 120 employees from Dissen produce gravure printing and embossing forms for printing packaging. Möller's company engraves motifs in around 20,000 printing rollers every year. At 1.1%, the error rate is already "very good", he says. But his target is 0.7%. This is because 92 faulty rollers per year amount to a six-figure sum. The work process is complex. Chromium and copper layers in the micrometer range are deposited and applied to the cylinders using electroplating, whereby various parameters determine the quality, such as pH value, current density, temperature, duration or degree of contamination of the bath. The baths are currently monitored manually and samples are sent to the laboratory for analysis.
Recognize sources of interference early
This is where the experts at the Green-AI Hub for SMEs come in. Additional sensors continuously monitor the baths and detect sources of disruption at an early stage. The artificial neural networks (ANN) analyze the data from the existing sensors in the machines and systems as well as the additional sensors installed in the baths to determine the quality of the gravure cylinders. The AI is supplemented by the use of computer vision, which automatically examines images of the gravure cylinders for defects and classifies the type of defect if necessary.
Large amounts of data are required to use the technologies, as well as initial manual work for training, as errors must be recognized and marked. After that, the data preparation step works automatically and corrections are only necessary in individual cases. 4Packaging expects an improvement of 20%, i.e. around 403 cylinders, which corresponds to a saving of at least half a million euros. Thanks to AI, the carbon footprint has also been reduced by 9 to 12 tons ofCO2 equivalents and the material footprint by 26 to 73 tonsof CO2 equivalents.
The data problem of SMEs
Jannis Vogel from the German Research Center for Artificial Intelligence (DFKI) worked as an AI expert at 4Packaging for months: "The problems faced by SMEs often start with building a suitable database for training the AI models. For example, sensor data is collected from a machine, but there is no link to the failure events." The requirements will become even more stringent over the next few years. In 2026, the first stage of the Ecodesign Regulation for Sustainable Products (ESPR) will take effect, initially for batteries and certain textiles, especially shoes, as well as
- Iron and steel, and later also electrical appliances, furniture and building materials. The Digital Product Passport (DPP) will play a central role in this. This is where all the data that is considered a prerequisite for a future circular economy will be collected:
- holistic transparency across the entire life cycle, including the procurement of raw materials
- ecological data on greenhouse gas emissions (product carbon footprint), water consumption and other environmental impacts
- ensuring the longevity, reusability, expandability and repairability of products
- all relevant information for recycling, such as ingredients, potential hazards, instructions for disassembly and proper disposal
How does the Digital Product Passport work?
Every player along the supply chain - from the raw material supplier to the manufacturer to the retailer - contributes to this information system. They are stored in a so-called registry. The DPP is based on the data that should already be available. The advantages are obvious:
- By accurately recording supplier information, materials, hazardous substances and production processes, a product can be better recycled and reused. This reduces the use of new raw materials and minimizes waste
- As the DPP contains detailed information about the materials, it becomes easier to repair or reuse used products. This leads to a reduction in resource consumption
- In the long term, the approach helps to reduceCO2 emissions and protect the environment. This is because products whose entire lifespan can be tracked transparently can be designed to be more environmentally friendly and better returned to the cycle
- Companies can access standardized and standardized data, which improves cooperation along the supply chain and enables efficiency increases
However, Alexander David from the start-up Greenable, which offers PCF and DPP services, knows that many SMEs are already overwhelmed by the last point. "The biggest challenge for companies is the internal data structures that have grown over the years. From paper documents and Excel sheets to ERP systems, the information is scattered everywhere. The DPP-relevant data must be available in a certain standardized form, for example machine-readable or interoperable, in order to be compatible with different IT systems along the supply chain. Then there is the semantic structure: how is acarbon footprint calculated or what does the recycled content of the material mean? No SME can do that on the side." They have to hire experts or buy in external expertise. "Of course, that costs money at first," David continues, "but it pays off in the long term."
The digital future is sustainable
Together with the Green AI Hub, Greenable presented the production island_Skye at Hannover Messe 2024, which is part of the SmartFactory Kaiserslautern demonstrator ecosystem. The focus was on how the DPP, which was implemented there in the administration shell, can open the door to the future of production. The _Skye checks recycled components and offers them for reuse, for example in data rooms. As soon as a component is searched for in the data room, _Skye offers it at a lower price and with the lowest possibleCO2 footprint. "This allowed us to demonstrate how the DPP helps to save resources and money," explains David. "In the long term, almost every sustainable measure has a positive economic impact."
SmartFactory Kaiserslautern, smartfactory.de










