Warehouse Planning Assistant: Integrating Simulation, Knowledge Graphs and LLMs

Analyzing large, complex output datasets from Discrete Event Simulations (DES) of warehouse operations to identify bottlenecks and inefficiencies is a critical yet challenging task, often demanding significant manual effort or specialized analytical tools. Our framework integrates Knowledge Graphs (KGs) and Large Language Model (LLM)-based agents to analyze complex DES output data from warehouse operations. It transforms raw DES data into a semantically rich KG, capturing relationships between simulation events and entities. An LLM-based agent uses iterative reasoning, generating interdependent sub-questions. For each sub-question, it creates Cypher queries for KG interaction, extracts information and self-reflects to correct errors. This adaptive, iterative and self-correcting process identifies operational issues mimicking human analysis. This approach for warehouse bottleneck identification, tested with equipment breakdowns and process irregularities, outperforms baseline methods involving modern reasoning models such as GPT-5. For operational questions, it achieves near-perfect pass rates in extracting accurate information. For complex investigative questions, we demonstrate its superior diagnostic ability to uncover subtle, interconnected issues. This work bridges simulation modeling, Knowledge Graphs and LLM reasoning, offering a more intuitive method for actionable insights and reducing time-to-insight. This generic framework can be readily adapted to other simulation-based planning tasks, including manufacturing and supply chain logistics.