TECHNOLOGY | INNOVATION
an entire production run, offering the potential for faster correction of deviations without the delay of lab testing and for identifying deviations that might be missed by a system that tests only selected samples.
As discussed in the August 2024 issue of
Above: At the University of Jena, synthesis robots are investigating the recycling of polymers
the Industry 4.0 vision for intelligent manufacturing, said Boqaileh. ML algorithms can be used to identify correla- tions between compounding parameters and product properties, but large quantities of data are needed, including metadata with details of the environmental conditions at time of data collection. While older testing equipment being used in the industry may not be capable of capturing or communicating this data, LabsCubed’s system is designed for Industry 4.0 and can be integrated with lab information management systems (LIMS) and manufacturing execution systems (MES). In the future, with such systems, data can be fed into AI/ ML models that predict potential quality deviations during the compounding process, and processing parameters can then be adjusted to maintain output quality.
Data analysis LabsCubed offers its AI Portal which uses a natural language interface to communicate data analysis results to users. Users can find trends or errors in data by asking the chatbot, for example, “Show me the highest stress value.” Other AI-based features in development will
further improve testing. For example, an AI-enhanced vision system inside the instrument can allow higher precision for elongation measurement. Predictive maintenance features will monitor machine health and automatically resolve minor issues. Boqaileh said that AI-based data analysis is finding use in areas in applications with many variables, including 3D printing as well as develop- ment of biodegradable compounds and com- pounds with recycled plastics content. He noted that high quality data is crucial: “If data isn’t good, the predictions aren’t good.” Spectroscopy-based in-line measurement systems can provide real-time data that monitors
14 COMPOUNDING WORLD | August 2025
Compounding World, ColVisTec is using its in-line fibre optic probes with ultraviolet-visible (UV-Vis), near infrared (NIR) and Raman-based spectroscopy to measure and control colour as well as identify impurities. These tools are particularly useful for compounding recycled materials, because they can detect changes in raw materials. Equitech International’s UV-Vis in-line spectro- photometer systems for colour measurement use a fibre optic probe inserted into the melt to provide real-time production data. After re-evaluating the technology last year for serviceability and manufac- turability and enhancing designs related to probes, spectrometer, electronics, and software, the spectrophotometer will now be capable of deliver- ing real-time results with lab-grade repeatability, as well as contain additional configurations and functionalities, the company announced. Last year, Equitech’s board initiated a compre-
hensive restructuring of operations. “Our goal remains to deliver advanced, cost-effective technol- ogies that not only rival traditional lab-based monitoring systems in performance, but also meet the price expectations of our customers,” the company said. A partnership with the University of North Carolina Charlotte (UNCC) “opens the door for future integration of emerging AI technologies from UNCC into Equitech’s product development pipeline”. In addition, a strategic agreement with M&M Technology gives Equitech access to scalable manufacturing space, skilled personnel, and engineering and technical resources to support future scaling efforts.
Recycled content The challenge of obtaining consistently good prop- erties in plastics compounds using inherently variable raw material streams from post-consumer recycled (PCR) plastics is an opportunity for AI-based solutions. Partners in the SpecReK project, with additional funding from the German Federal Ministry of Education and Research (BMBF), are investigating how to use real-time testing of recycled materials combined with AI models to improve products. The collaborators include BASF, Endress+Hauser, TechnoCompound, the University of Bayreuth and Friedrich Schiller University Jena. “We do not currently have the necessary analysis
www.compoundingworld.com
IMAGE: ANNA SCHROLL
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50
