KIOptiPack - Holistic AI-based optimization of plastic packaging with recyclates content

  Copyright: © WZL  

The aim of KIOptiPack is to provide and validate practice-ready AI-based tools for successful product design as well as quality-assured production of plastic packaging with a high amount of recycled material. The tools shall be provided in an AI application and data space and shall be transferred to application in industry. The tools and data space are connected to a central network platform for value engineering. The project and the tools developed take the entire value chain including secondary raw materials, material and packaging development, process design, packaging production to waste collection and processing into account.

The following RWTH institutes are involved in the project: WZL, TIM and IKV. The WZL-IMA is developing an infrastructure for the life cycle management of machine learning models. This includes the development of standardized pipelines for model development, validation and deployment. In addition, the infrastructure will be transferred to the application domain by developing an overarching system for the manufacturing process of plastic packaging. Furthermore, the WZL-IMA supports, among others, the development of models for the film layer structure by creating hybrid simulations as well as the sustainability assessment.

WZL-MQ is working on the holistic sustainability assessment of the overall system. For this purpose, processes and business model requirements as well as customer requirements for sustainable packaging are determined. These are coupled with the results of the objective sustainability assessment and process optimization in order to integrate knowledge about customer acceptance of sustainable solutions into process optimization and to increase the acceptance of sustainable plastics packaging by the customer.