(90b) Transforming Chemical Manufacturing Decision Making with Retrieval Augmented Generation (RAG): Grounding AI Responses with Domain Specific Data

Facing the contemporary challenges in the chemical manufacturing sector, the integration of advanced computational technologies becomes more critical than ever. This presentation aims to discuss the novel technique of Resource Augmented Generation (RAG), a state-of-the-art that transforms how generative AI can formulate responses without the necessity for expensive retraining or fine-tuning of Large Language Models (LLMs).

RAG is an emergent solution that bridges the gap between AI capacity and specialized knowledge, by grounding AI-generated responses on domain specific data sources. Our method integrates RAG into strategic decision-making workflows by aligning it with our existing databases, thus effectively leveraging AI's potential to produce contextual responses on otherwise unfamiliar subjects.

We intend to share real examples of how RAG is used in our chemical manufacturing business' strategic planning. Even at the early stages integrating RAG boosted GenAI into everyday decision-making processes, has improved our overall efficiency and allowed us to make more informed decisions. It's our hope that these cases will showcase how leveraging GenAI and RAG can have a tangible impact on chemical manufacturing strategies.