AI for Chemical and Biological Manufacturing

The workshop highlighted the effort towards scalable and sustainable processes, propelled by advancements in AI and machine learning. AI-driven process optimization and material’s discovery was discussed across different scales. It was noted that from molecular discovery to reactor-scale operations, AI is accelerating the understanding of complex process landscapes, enabling multi-objective optimization, and facilitating image-based measurements for enhanced process control.

In biological manufacturing, AI was discussed as a means of revolutionizing bioprocessing through automated quality checks, cell media optimization, and design of experiments, promising significant improvements in efficiency and yield.

In chemical manufacturing, sustainability is paramount, with AI being leveraged to reduce carbon footprints and transition towards circular economy principles. The talks highlighted the pivotal role of AI in discovering sustainable chemical routes, optimizing processes, and enhancing control systems. Digital transformation and modularization emerged as key strategies to streamline operations, with integrated digital twins enabling comprehensive monitoring and management of process plants. A fully integrated AI system would allow researchers and engineers to efficiently test and optimize new technologies while ensuring adaptability to evolving demands.

The challenges of building and accessing open-source, diverse datasets to enhance ML models, balancing domain knowledge with real-world data inputs, and scaling digital twins from lab to manufacturing levels were highlighted. The difficulty of ensuring the manufacturability of biologic products, considering factors like time, cost, and sustainability, and addressing data limitations and noise was also noted.