Sustainable plastic packaging: AI supports design and production
15.03.2023
Plastic packaging made from recycled material? Yes, but please as high quality as possible – and recyclable! This poses great challenges for manufacturers, because the use of recycled materials still often leads to downcycling. One key to more sustainable plastic packaging lies in digitalisation. This is also where the KIOptiPack project comes in.
In an interview with K-MAG, Mauritius Schmitz talks about the requirements for plastic packaging, the role that design plays in the sustainability of packaging and the extent to which artificial intelligence can support this.
To make plastic packaging more sustainable, the entire value chain must be considered. The KIOptiPack innovation lab deals with the beginning of the value chain - design and production. What role does design already play in sustainability?
Mauritius Schmitz: Product design plays a central role in the sustainability of products. It defines the choice and consumption of materials and thus the types and quantities of plastic introduced into the cycle. It also influences functionality and suitability for (packaging) use, manufacturability and production efficiency. Furthermore, product design can facilitate or complicate the sorting, separation and reprocessing processes – which in turn determine the quality of the resulting recyclates.
Exhibitors & products around recycling
- Recycling plants
- Separating and sorting systems for waste
- Size reduction equipment (crushers, shredders, grinders)
- Compounding lines
- Recyclates
How can AI be used in the design of packaging?
Schmitz: Even with sustainable packaging design, the same, or at least comparable, expectations and criteria must be fulfilled for the product as for products made of virgin material.
This is about product functionalities, such as barrier properties, purity and mechanical stability, and about subjective requirements from the consumer's point of view, such as appearance, smell and safety. But also the processability of the chosen material in the production process plays a decisive role for a functioning product design, by this I mean, for example, the thermal or flow properties of the plastic.
AI can be used to take these complex and diverse influences into account and make them optimisable. By using AI, these influences can be made quantifiable and packaging designers can get a better idea of how their designs affect the different requirements and how they can optimise them to create more sustainable packaging.
How can AI make production more sustainable?
Schmitz: To improve the quality of plastic packaging, AI is used in different production phases. AI is already used at the material receiving stage to detect material fluctuations and adjust the processing accordingly. During the processing stages, neural networks are used to generate and optimise process models. In this way, correlations between material characteristics, process settings and quality can be recognised and used for correct process adjustment. In the KIOptiPack project, the processing methods compounding, flat film and blown film extrusion, injection moulding and thermoforming are specifically considered.
Why is cooperation with industry – i.e. companies like Brückner, Reifenhäuser or Pöppelmann – important?
Schmitz: Cooperation with industry is of great importance and depends in its way on the respective industry. However, an important aspect that has not been sufficiently considered so far is the consideration of the entire value chain. In order to achieve optimal sustainability, changes have to be made at various points in the value chain, which may not be initiated and implemented out of their own motivation. For example, the responsibility for more sustainable plastic packaging could be shifted to the collection and reprocessing process alone, but this would certainly not be the most effective and sustainable approach. Instead, it makes more sense to develop incentives and business models that enable global change and load-sharing along the entire value chain.
To achieve this goal, it is essential to understand the hurdles and potentials for each participant. Through the AI Application Hub, which networks project partners from established machine manufacturers to technology start-ups, as well as from research institutes to food manufacturers and distributors, the possibility of a direct exchange along the entire value chain is created. Only in this way can we achieve a common understanding of the potentials of more sustainable packaging production.
What will the plastic packaging of the future look like?
Schmitz: It is difficult to predict what packaging will look like, because we will also be able to develop and produce sustainable plastic packaging in great variety in the future. These will be individually tailored to the specific application, so that they fulfil minimal material consumption and at the same time reliable suitability for the application. At the same time, however, this packaging will have to be returned to the cycle with the minimum possible losses.
What can already be predicted with great certainty, however, is that the packaging of the future will contain a maximum proportion of recyclate. It will preferably contain only one type of polymer and be easier to handle when sorting and separating. In order to improve and simplify the sorting processes, digital – and at best global – product or material passports, which are currently being developed through various initiatives, will play a decisive role. And in this area, we will not make progress without digitalisation and artificial intelligence methods.
The joint project KIOptiPack with around 40 project partners is funded by the Federal German Ministry of Education and Research (BMBF) as part of the "Research for Sustainability" ("Forschung für die Nachhaltigkeit", FONA) initiative and, together with the partner laboratory K3I-Cycling, maps the entire cycle for plastic packaging. More information here.