Agentic ERP: Multi-Agent Large Language Model Architecture for Autonomous Enterprise Resource Planning
Zhihao Liu, Tianyu Wang, Xi Vincent Wang, Lihui Wang
Why It Matters
What makes this one worth your time
This work is relevant for AI researchers and engineers interested in applying LLMs to automate complex enterprise systems, potentially reducing human workload and improving operational efficiency.
Agentic ERP leverages multi-agent LLMs to autonomously manage ERP systems with human oversight.
Summary
The paper introduces Agentic ERP, a multi-agent large language model architecture designed to autonomously manage enterprise resource planning systems by integrating role-aligned agents with human oversight and a graph-based orchestration system. The approach is evaluated through scenario-based tasks, orchestration paradigm comparisons, and a year-long simulation, demonstrating improved performance over rule-based systems.
Key contributions
- Introduction of a multi-agent LLM architecture for ERP systems.
- Development of a graph-based Planner--Executor--Reflector--Responder orchestration.
- Comprehensive evaluation of the system against rule-based and no-intervention baselines.
Notable insights
- Role-aligned agents reduce tool-selection complexity in ERP systems.
- Graph-based orchestration decouples generation from evaluation using external grading criteria.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2607.17331v1 Announce Type: new Abstract: Enterprise Resource Planning (ERP) systems record transactions reliably but still delegate almost all operational decision-making to human specialists, because classical rule-based automation cannot reason about exceptions and monolithic AI assistants degrade when asked to coordinate across functional boundaries. This paper presents Agentic ERP, an expert-system architecture that combines role-aligned large-language-model (LLM) agents with a risk-tiered human-in-the-loop harness and a graph-based orchestrator to execute end-to-end business workflows on a production ERP backend. First, autonomous ERP operation is formulated as a constrained sequential-decision problem over a structured enterprise state, with a decomposition argument linking role-aligned agents to a measurable reduction in per-step tool-selection complexity. Second, a graph-based Planner--Executor--Reflector--Responder orchestration decouples generation from evaluation through externalised grading criteria and sprint contracts, packaging recent harness-engineering principles as inspectable expert-system artefacts. Third, the system is evaluated at three levels: a scenario-based task suite, a comprehensive comparison of six orchestration paradigms on cross-functional crisis tasks, and a 365-day agent-in-the-loop simulation against rule-based RPA and no-intervention baselines. Across these levels the proposed multi-agent method is significantly better than the baseline, and the system sustains a simulated year of operation with zero stockouts while the rule-based baseline accumulates hundreds under the same demand stream. The work shows that role-aligned LLM agents under human oversight can move an ERP system from passively recording transactions to actively executing operational decisions, and it provides a reference architecture and an evaluation protocol for autonomous enterprise resource planning.