Quality Injection: Development of an injection molding process assistance system for zero-defect production, consisting of AI-based quality prediction and control parameter prediction system.

The aim of the project is to develop an AI-based injection molding assistant to enable zero-defect production. This assistant consists of two AI systems: A quality prediction system and a control parameter prediction system.
The quality prediction system is intended to drastically reduce or eliminate the need for quality assurance checks. For this purpose, pressure and temperature sensors are primarily installed in the injection mold and data is read out during the process. Based on this data, the AI-based quality prediction system should predict whether the component meets the tolerances. If the molded components are out of tolerance due to batch variations, for example, another AI system is to be used within a few cycles to suggest machine parameters that can be used to produce components that meet tolerances again. Here, machine parameters must be found which, on the one hand, detect the loss of quality and, on the other hand, can also be readjusted. The use of recycled components in the plastic granulate is the main focus here in order to comply with possible legal regulations. 

Duration: April 2023 - March 2025
Project partner: Hanselmann & Cie. technologies GmbH
Grant reference: BW1_3060/02
Project sponsor: investBW
Contact persons: Prof. Dr.-Ing. Matthias Deckert, Elena Fischer, M.Sc.