Enhancing Data-Driven Decision-Making in Chemical Manufacturing Industry: A Hybrid Predictive and Generative AI Decision Logic, Human-Centric Autonomous Workflow

In the rapidly evolving landscape of the chemical manufacturing industry, the integration of artificial intelligence (AI) into decision-making processes is becoming increasingly essential. This talk will delve into the development and implementation of a hybrid generative and predictive AI workflow that leverages data driven, and yet physics-informed predictive models as well as a context-aware generative ai workflow. We will explore how this system operates when the “fast, System 1 thinking” like predictive AI encounters high uncertainty, triggering a “slow, System 2 thinking” like generative AI to engage in a multi-step reasoning and resolution approach, and a final comparative analysis to determine solution confidence levels.

Furthermore, the talk will highlight the symbiotic relationship between AI and human expertise, emphasizing the workflow where AI autonomously tackles problems and seeks human intervention upon reaching a low confidence threshold. We will discuss the role of multi-modal AI in complex document understanding, showcasing how it aids in deciphering diverse data formats within the chemical manufacturing domain.

Lastly, we will examine the agentive workflow, where AI acts as an autonomous agent orchestrating the overall workflow, and its implications for efficiency and innovation in the industry. The talk aims to provide insights into how these advanced AI workflows can be tailored to enhance decision-making, optimize operations, and maintain a competitive edge in the chemical manufacturing industry.