Multi-agent systems for chemical engineering: A review and perspective

2025-08-11

AI Summary

To address the challenge of intelligent decomposition and collaborative optimization of complex chemical process workflows, this study proposes a domain-specific large language model (LLM)-driven multi-agent system (MAS) architecture. Methodologically, it introduces a role-specialized agent coordination mechanism, integrates heterogeneous multimodal chemical engineering data—including process flow diagrams, thermophysical property databases, and experimental logs—to construct a domain-adaptive foundation model, and incorporates tool-calling capabilities with interpretable decision-making modules. The key contributions are: (i) the first systematic establishment of an MAS collaboration paradigm tailored to chemical engineering, enabling end-to-end support from task parsing and process reconfiguration to autonomous decision-making under safety and environmental constraints; and (ii) significant improvements in accuracy and traceability for process simulation, fault diagnosis, and optimization design—providing a scalable theoretical framework and empirical pathway toward intelligent chemical process systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Coordination and Collaboration
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Large language model (LLM)-based multi-agent systems (MASs) are a recent but rapidly evolving technology with the potential to transform chemical engineering by decomposing complex workflows into teams of collaborative agents with specialized knowledge and tools. This review surveys the state-of-the-art of MAS within chemical engineering. While early studies demonstrate promising results, scientific challenges remain, including the design of tailored architectures, integration of heterogeneous data modalities, development of foundation models with domain-specific modalities, and strategies for ensuring transparency, safety, and environmental impact. As a young but fast-moving field, MASs offer exciting opportunities to rethink chemical engineering workflows.

Problem

• Decomposing complex workflows into collaborative specialized agents
• Integrating heterogeneous data modalities in multi-agent systems
• Ensuring transparency, safety, and environmental impact in MAS

Innovation

• LLM-based multi-agent systems for workflows
• Specialized agents with domain knowledge
• Integration of heterogeneous data modalities