We’ll be speaking at this one-day conference focused on advances in coating technology which, this year, has a special focus on AI in formulation.
Look out for Dr Bogdan Nenchev’s presentation in the scientific sessions.
Talk details
Beyond Trial and Error: Machine Learning for Accelerated Coating Formulation
Formulation development for coatings is a combinatorial challenge; the number of potential ingredient combinations, proportions, and processing conditions can run into the thousands, far outpacing what conventional experimental programmes can explore. Machine learning (ML) has emerged as a powerful tool to navigate this design space, complementing chemist intuition and experience to focus experimental effort where it matters most.
Real-world deployments are already delivering measurable impact. At FUCHS, the world’s largest independent lubricant manufacturer, Alchemite™ ML was applied to adapt an existing formulation to a new market segment. Training on just 24 historical formulations, the model guided three cycles of adaptive experimental design, proposing 30 candidates and ultimately delivering a 50% improvement in key performance metrics – in a fraction of the time a conventional approach would have required. Applications have spanned pipe coatings, corrosion-protection coatings, hydraulic fluids, and metal working fluids.
The FUCHS case demonstrates what is possible at the level of a single project; the broader opportunity lies in applying this approach systematically across product portfolios and R&D pipelines throughout the coatings industry.
