Building Automotive Parts Intelligence using Graph Memory for Supply Chain Resilience

This video demonstrates a dual-store system using Neo4j for graph reasoning and Qdrant for semantic search, orchestrated by a LangGraph agent to analyze supply chain risks in automotive parts, validate substitutes, and assess supplier concentration.

What it covers

Supply Chain Risk in Automotive Parts
A chip shortage causes a critical part to become unavailable, highlighting vulnerabilities in automotive supply chains.
Vector Search Limitations
Vector search can identify similar parts but fails to determine affected vehicles, compliance status, or supplier risk.
Dual-Store Architecture
The system combines Neo4j for graph-based reasoning and Qdrant for semantic search to enhance part intelligence.
LangGraph Agent Orchestration
A LangGraph agent coordinates impact analysis, compliance validation, and supplier risk assessment across the supply chain.
Impact Analysis and Substitution Validation
The agent evaluates which vehicles are affected and verifies that substitute parts meet BOM and compliance requirements.
Supplier Concentration Risk
The system identifies and flags risks arising from over-reliance on single suppliers.
Agentic AI in Supply Chain Resilience
The integration of graph memory, vectors, and agents enables real-time, intelligent decision-making under supply chain pressure.