Skip to main content
Back to challenges
Challenge

Manufacturing Knowledge Graph RAG System with LlamaIndex

Factory managers struggle to query disparate MES logs, maintenance records, and quality documents simultaneously. Build an enterprise manufacturing RAG pipeline using LlamaIndex to query structured asset data and unstructured downtime reports. Reach an F1 score above 0.88 on complex operational query benchmarks.

EngineeringHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build an enterprise manufacturing search engine with LlamaIndex connecting structured SQL tables and unstructured equipment manuals.

Delivery guide

How work is evaluated

Evaluation

Evaluate retrieval precision and answer correctness across complex plant operational queries.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

citation_correctness

Cites correct source document and table

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

f1_score

F1 evaluation score on factual operational QA • target: 0.88 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Construct a LlamaIndex PropertyGraphIndex combining unstructured manual PDFs and structured SQL MES logs

  • Implement hybrid retrieval (vector search + graph traversal) for manufacturing operations

  • Evaluate response accuracy against factory operational query ground truths

  • Optimize indexing speed and query latency for large manual sets

Resources and assets

Reference links and supporting material

Dataset notes

Corpus of factory downtime records, work orders, and PDF technical service manuals.

Challenge setup

Evaluate retrieval precision and answer correctness across complex plant operational queries.

Challenge input

JSON object with query string

LlamaIndex

Framework for data indexing and graph rag operations.

Evaluated output

JSON object with answer, source_nodes, and metric_citations

Evaluation checks
  • Cites correct source document and table
  • F1 evaluation score on factual operational QA • target: 0.88 • range: 0-1
Configured evaluation evidence
  • Benchmark: Enterprise RAG Industrial Benchmark
  • F1 Score target: 0.88
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Llama Index · Required
  • LlamaIndex · Optional
  • Llama 3.2 · Optional

Evaluation contract

  • citation_correctness · Weight 1
  • f1_score · Weight 1

Metrics

  • f1_score · Target 0.88

Benchmarks

  • Enterprise RAG Industrial Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with query string

    Output: JSON object with answer, source_nodes, and metric_citations

    Execution harness: Python

Evaluation tools

  • LlamaIndex · RAG Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Manufacturing Knowledge Graph RAG System with LlamaIndex