Accelerating innovation for plastics, elastomers, paints and other polymeric products.
Polymer R&D supports applications in almost every industrial sector. There is a continued need for fast innovation, not only to find performance improvements, but also to address challenges including cost and energy price volatility, sustainability, regulatory constraints on ingredients and processes, and supply chain disruption. Achieving speed and resilience requires teams to understand complex, high-dimensional, non-linear relationships in polymer formulations in a context where key property data is often incomplete. The result is high reliance on costly experimentation.
Machine learning can help by learning from experimental and process data to create models that capture these high-dimensional, non-linear relationships. Such models can then be used to explore polymer formulation space, to predict and test potential solutions, and to focus experimental work. But machine learning methods can struggle where data is sparse, as it often is in polymer R&D. And these methods can also be difficult for polymer domain experts to apply and interpret. We introduce Alchemite™, machine learning software that overcomes these difficulties. We discuss an application example in which Alchemite™ has been used to reduce experimental workloads and provide valuable insight into polymer blending. Similar reported applications include lubricants, paints, inks, and materials design.