Fraunhofer Institute for Process Engineering and Packaging IVV
Fraunhofer Institute for Process Engineering and Packaging IVV
KI Optipack

AI APPLICATION HUB ON PLASTIC PACKAGING – Sustainable Circular Economy through Artificial Intelligence

KIOptiPack: an Innovation Lab of the BMBF funding measure AI Hub

A more sustainable circular economy for plastic packaging

Plastics are one of the most versatile and popular materials for packaging and would also be put to good use in terms of the bioeconomy if their secondary raw materials were recycled in higher proportions. For this reason, more and more voices are being raised in European and German politics in support of the idea of the circular economy and the need to increase the use of recycled materials. Currently, however, technical and economic challenges still stand in the way.

Which technical means are needed and which economic challenges have to be overcome in order to bring the use of recycled materials into the cycle in a meaningful and cost-efficient way?

This is precisely where the funding measure "AI Application Hub on Plastic Packaging", funded by the German Federal Ministry of Education and Research (BMBF) comes in. The aim is to pave the way for a more sustainable circular economy through the use of artificial intelligence.

The "AI Application Hub on Plastic Packaging"

In order to close the cycle for plastic packaging as far as possible, 51 partners from industry, science and society are working together in two innovation labs: KIOptiPack for design and production and K3I-Cycling for cycle closure. They were set up to enable cross-laboratory data exchange and to ensure that all relevant findings across the entire value chain are also considered.

Logo AI Hub Plastic Packaging
Logo K3I Cycling
Logo KI Optipack

So far, neither the quality nor the degree of purity of secondary raw materials can be predicted or defined correctly.  The main reasons for this are, for one, the impurities resulting from the substances in the material cycle and, for another, changes in the polymer and functional properties of the recyclates that occur after recycling. This leads to changes and variations in the processing and use of the secondary raw materials, which in turn result in reduced quality and efficiency in the production process. In addition, their legal compliance for food and packaging safety cannot yet be guaranteed. To avoid quality losses and complications, recyclates are therefore still used far too little.

»Design and production« - the Innovation Lab KIOptiPack

The goal of the KIOptiPack Innovation Lab is to roll out, validate and put into application AI-supported tools for successful product design and quality production methods for plastic packaging made from a high proportion of recycled materials. In addition, a central network platform for value creation engineering will be formed for this purpose and linked to the AI application and data space. To this end, our team at the Fraunhofer IVV brings its interdisciplinary expertise from the fields of packaging development and manufacturing, food science, sensory analytics, measurement technology, AI methods and algorithms.

KI OPTIPACK interdisciplinary expertise of the Fraunhofer IVV in the
BMBF funding measurement "AI Application Hub on Plastic Packaging"

Our experts combine years of research and knowledge from the evaluation of packaging materials in accordance with food legislation to the production of finished packaging. In addition, they have the necessary digitization solutions required in the packaging sector. Find out what our researchers focus on and how our interdisciplinary team works together in the Innovation Lab to make the plastic packaging cycle more sustainable.

Our part in the Innovation Lab KIOptiPack

With the aim of functional material use of recycled polymer grades in packaging manufacturing processes, we are linking all material data along the entire value chain for the first time:

 

  • Contaminants and foreign molecules from recycling
  • Odorous substances
  • Decisive parameters for describing the processing behavior of plastics and films
  • Guideline criteria for processing success
  • Machine configuration and settings

 

Packaging development

Sustainable packaging systems

We research the optimization of recyclable and recyclate-containing packaging systems:
 

Packaging production

Packaging processes

We research the optimized processing of recyclable and recyclate-containing materials into packaging: 

 

© iStock.com / metamorworks

Packaged foods

Quality & shelf life

We research spoilage processes as basis for customized measures to prolong the shelf life of packaged foods:

 

  • Product-specific storage tests for shelf life evaluation
  • Derivation of packaging requirements
  • Mathematical models for rapid shelf life prognosis
  • Customized concepts for prolonged shelf life

© Dvoinik / iStock.com

Odor optimization and characterization

Sensory analytics

We study the effect of packaging containing recyclates on consumers:

 

© iStock.com / artisteer, iStock.com / amtitus

AI development for recycling processes

Efficient algorithms

We develop tools for AI-assisted prediction of the processability of materials as well as the manufacturability of packaging:

  • Chemical-instrumental identification of undesired odorous substances using automated data analysis techniques
  • odor prediction using known substance lists with the aid of AI
  • Collection of data on the influence of different recyclate classes, sources and qualities of polyolefins on the processability and manufacturability of packaging systems

AI-based optimization of packaging: the approach of the KIOptiPack Innovation Lab

In this project, the research team will map the entire value chain of the packaging industry and include data and findings from secondary raw materials, material and packaging development, process design and packaging production as well as the consumers.

For this purpose, a data infrastructure will be formed in which the necessary data will be made available and in which the AI-supported tools will be developed.

Furthermore, agile tools will ensure a continuous analysis and observation of the qualification of the materials and their allocation. The goal is to detect and predict undesirable changes in a timely manner. At the same time, adaptive-controlled AI assistance systems link the relevant information and adjustment instructions, transmit them to the machines and communicate suggestions for target-oriented process changes to the operators. The aim is to increase the quality, robustness and productivity of packaging materials containing recycled materials.

The provided data and recommendations for action will ultimately be merged with further data in an AI-supported packaging development system. This will create a data pool of current raw material data, availabilities, process limits as well as packaging requirements. Thus, for the first time, the production of packaging with maximum recycled content in the required minimum quality will be possible.

The desired transparency and the exact assessment of the quality and purity of the recyclates create the basis for an efficient use of the material flows in the cycle.

Milestones

Project steps

 
  • Establishing a data room and providing a data infrastructure
  • Development of new analysis tools for data reduction and filtering
  • Standardization of data semantics and development of an ontology
  • Development of new process optimization models in material processing
  • System linkage of the models and development of a data hub 
  • End-to-end objective and subjective sustainability assessment using life cycle assessment methods
  • Business model and ecosystem development

© iStock.com / Sandipkumar Patel, iStock.com / kolotuschenko

Demonstrator 1

Packaging example

Develop a packaging example that will demonstrate the viability of processing operations related to materials containing recyclates.

Demonstrator 2

Data interface

Development of the first data infrastructure with data interface for connecting existing databases of companies, production machines as well as data resulting from AI-based models.

Demonstrator 3

Recyclability

Development of a system that combines the results of the previous data processing and both assesses and provides them with regard to recyclability in accordance with the assessment system recognized by the Packaging Register.

Project management / funding:

Projektträger Jülich GmbH PTJ / Federal Ministry for Education and Research BMBF

Logo Federal Ministry of Education and Research BMBF
Logo: FONA Research for sustainability
Logo: AI National strategy for artificial intelligence

More information

 

The AI Application Hub on Plastic Packaging

Easy packaging

Learn how Artificial Intelligence (AI) can help reduce piles of waste and recycle important resources.

More

Our services