Coatings Intelligence: Data-driven Automation for Chemistry 4.0

Organic coatings and paints play a vital role in safeguarding surfaces, but improving their properties typically requires a large number of carefully planned experiments. High-throughput screening (HTS), employing industrial robots, allows automated preparation and testing, thus speeding up research and development on materials. Challenges remain in hardware interfaces, data management, and plant control system adaptation. Machine Learning (ML) is increasingly applied to various applications in chemistry for experiment suggestions and analysis; thereby paving the way for a higher degree of automation in HTS. This work presents an integrated system composed of an automation platform to perform HTS, a characterization module to evaluate the material, and a DL model to accelerate the development of new coatings by evaluating their scratch resistance using sensor data independent of images. The model is trained on data managed by a data lake. This integration showcases the interplay between chemistry, automation, and data science as a basis for smart manufacturing.