Simulation, Machine Learning, and Optimization in the Coatings Laboratory

 
Automation and robotics are becoming common technologies in the coatings industry. These technologies, which are disrupting material innovation and product development workflows, still ultimately rely on human intuition, decision making, and experiment design methodologies. As experimentation and testing becomes increasingly automated and resource-efficient, a commensurate improvement in the application of computational design tools is needed. These include simulation, machine learning, and multi-objective optimization methods. This talk will present updates on the incorporation of multiscale modeling, deep learning neural networks, and evolutionary optimization methods into coating development and qualification processes, with an emphasis on making decisions under uncertainty, overcoming barriers to implementation, and countering the culture change roadblocks that are often experienced across large organizations and cross-functional teams.