WarehouseAI: A Closed-Loop Multi-Agent Framework for Industrial Operations Planning

Enterprise AI agents often struggle to accurately reason over dense, multi-stage relational logs produced by industrial simulators. We introduce WarehouseAI, a closed-loop multi-agent system for automated warehouse optimization built on a Knowledge Graph-grounded framework. The architecture translates natural language intent into simulation parameters, executes a SimPy Discrete Event Simulation (DES), and maps the resulting event logs into a Neo4j knowledge graph. A dual-path Knowledge Graph Question Answering (KG-QA) reasoning chain processes operational queries through direct Cypher retrieval and investigative queries through adaptive multi-hop evidence accumulation, with semantically decomposed query generation and two-level self-correction. Six specialized agents, coordinated hierarchically, execute a complete planning cycle: Configure $\rightarrow$ Simulate $\rightarrow$ Analyze $\rightarrow$ Recommend $\rightarrow$ Validate. Across 9 independent trials spanning 3 bottleneck scenarios with non-obvious root causes, the system achieves 100\% intent-routing accuracy, simulation validity, recommendation grounding, and loop improvement rate, with composite scores of 0.780-0.859 and an average 15.0\% reduction in package processing time per cycle. The KG-QA engine reaches Pass@1 = 0.92 on a 25-question benchmark substantially outperforming single-pass generation (0.59) and self-reflection (0.70). Results show that structural grounding and hierarchical agent orchestration together deliver the reliability and efficiency required for autonomous industrial decision-making.