Multi-Agent LLMs for Automating Sustainable Operational Decision-Making

Operational decision-making in Process Systems Engineering (PSE) has achieved high proficiency at specific levels, such as supply chain optimization and unit-operation optimization. However, a critical challenge remains: integrating these layers of optimization into a cohesive, hierarchical decision-making framework that enables sustainable and automated operations. Addressing this challenge requires systems capable of coordinating multi-level decisions while maintaining interpretability and adaptability. Multi-agent frameworks based on Large Language Models (LLMs) have demonstrated significant promise in other domains, successfully simulating traditional human decision-making tasks and tackling complex, multi-stage problems. This paper explores their potential application within operational decision-making for PSE, focusing on sustainability-driven objectives. A realistic Gas-Oil Separation Plant (GOSP) network is used as a case study, mimicking a hierarchical workflow that spans from initial back-of-the-envelope multi-objective optimization for cost-emissions trade-offs to a negotiation phase reflecting upper management decision-making, and culminating in high-fidelity simulations to validate operational setpoints at the plant level. This workflow serves as a canvas to assess the benefits of multi-agent LLMs, including their ability to integrate multi-layered decisions, enhance the explainability of strategies, and streamline automation in PSE workflows. The results demonstrate the potential of multi-agent LLMs to address the integration challenge in PSE, supporting sustainable and efficient operational decisions. Beyond GOSPs, this research highlights promising applications of multi-agent LLMs across process engineering, contributing to the vision of hierarchical, automated decision-making for the �plant of the future,� where diverse models and tools operate within an intelligent, unified framework.