09.10.2024 by Dr. Marc Egelhofer
AI in the Laboratory: Optimized Data Analysis and Quality Control of Plastic Recyclates
Laboratories are often confronted with an overwhelming amount of data from different sources and instruments. Managing, analyzing and harnessing this amount of data is a challenge, especially when the data is siloed across different systems. This is particularly true for the quality control of polymer recyclates in plastics processing. DIN SPEC 91446 defines four quality levels for recyclates, which depend on the quality of the data. The higher the quality level, the more quality-relevant data must be collected.
For the laboratory, maximum quality therefore also means a maximum number of data points that must be analyzed and documented in daily practice. One of the key questions for laboratories therefore is how modern technologies such as artificial intelligence (AI) can be integrated into existing processes in order to increase data quality and efficiency.
AI - A Polymer Recyclate Use Case
One case study involves a plastics processor procuring recyclate mixtures from various suppliers. The company was confronted with highly fragmented data management. Measured values required for quality control were scattered across various Excel spreadsheets, and manually analysis of this data took up considerable resources.
In such an environment, the use of AI cannot provide any significant relief. It was only by introducing a data-based platform in combination with an AI-supported digital assistant that the plastics processor was able to significantly optimize his processes. The platform integrated all relevant device data, thus providing a complete overview of the materials being analyzed.
The AI-supported digital assistant can then unleash its full potential − from complex data analysis and identification of correlations to data visualization and report generation.
Supplier Monitoring Scenario
A particularly illustrative scenario shows how supplier monitoring is made easier with the help of AI. The case study shows, for example, how a simple query was used to analyze data from three different suppliers and quickly identify deviations in the batches supplied. In one case, AI identified that one of the suppliers had delivered a batch with three times the viscosity agreed. This precise data analysis enabled the company to better assess the quality of suppliers and make informed decisions about future suppliers.
It was also possible to avoid using a material with too high a viscosity in advance, which could have led to major processing problems and lower product quality.
Conclusion
This example from a case study shows that AI is no longer a vision of the future, but is already being used in laboratories today. It helps efficiently analyze large amounts of data, thereby ensuring the quality of recycled plastics. Laboratories that rely on AI-supported data management solutions not only benefit from improved data quality, but also from considerable time and resource savings. This ultimately increases competitiveness and helps develop better products.
Read the full case study for more examples of how AI can support the lab: