SupplyLens

In 2019, Volkswagen committed to making electric vehicles more sustainable.
They published ESG reports, used certified suppliers, and pledged clean sourcing.

But a 2020 investigation by Amnesty International found:

• ❗ Cobalt in VW’s batteries traced to mines in the DRC
  
• ❗ These mines used child labor and unsafe conditions
  
• ❗ The violations were several tiers deep — invisible to VW’s standard audits

🔶 Tier 1 looked clean.
All certifications were in place.

🔴 Tier 3 was the problem.
Child labor and unsafe mines were hidden beneath the audit trail.

“The certification meant nothing.
We didn’t see where the real risk was.”
Internal source, Amnesty report (2020) —

Source:
Amnesty International, “Powering Change: The supply chains behind clean cars” (2020)
https://www.amnesty.org/en/documents/afr62/7395/2017/en/

In 2020, Nestlé pledged to eliminate deforestation from its supply chains and committed to sourcing sustainable palm oil.

However, investigations revealed:

• ❗ Nestlé continued sourcing palm oil from mills linked to deforestation and human rights abuses.
• ❗ Satellite monitoring detected over 1,000 deforestation alerts per day near Nestlé’s palm oil suppliers.
• ❗ Reports indicated worker deaths, injuries, and land rights violations associated with some of Nestlé’s palm oil sources.

Despite public commitments, these issues persisted in Nestlé’s supply chain.

🔶 Public Commitments vs. Reality
Nestlé’s sustainability pledges did not align with on-the-ground practices.

🔴 Ongoing Deforestation
Continuous forest clearing was detected near supplier mills.

❌ Human Rights Concerns
Allegations included worker exploitation and land rights violations.

Sources:

• Palm Oil Detectives, Nestlé, 2021. Link
• SwissInfo, Nestlé identifies over 1,000 deforestation alerts per day, 2020.

In February 2018, KFC UK faced an unexpected crisis: a chicken shortage. The fast-food chain had recently switched its logistics partner from Bidvest, which operated a network of six warehouses, to DHL, which centralized operations to a single warehouse in Rugby.

This strategic change led to:

• ❗ Over 750 KFC outlets across the UK and Ireland temporarily closing due to delivery failures.
• ❗ Significant food waste and financial losses.
• ❗ Public relations challenges, as customers expressed frustration over the lack of chicken.

The incident highlighted the risks of over-centralization and inadequate contingency planning in supply chain management.

🔶 Centralized Logistics Failure
Transitioning to a single warehouse created a single point of failure.

🔴 Operational Disruptions
Delivery issues led to widespread store closures and supply shortages.

❌ Brand Reputation Impact
The shortage became a national news story, affecting customer trust.

Source:
Wired, “The inside story of the great KFC chicken shortage of 2018” (2018)
https://www.wired.com/story/kfc-chicken-crisis-shortage-supply-chain-logistics-experts

• 94% of companies have no visibility beyond Tier 1
  – Deloitte, 2022

• Up to 90% of ESG violations happen where data disappears**
  – McKinsey, 2021

• CSRD, CS3D, EUDR demand real due diligence
  – Risk mapping
  – Documented origin
  – Verified suppliers

Supply chains don’t just suffer from a lack of data.

They suffer from a lack of trust — and incentives to hide.

Every player has a reason to stay quiet:

• 📦 Tier 1 suppliers don’t want to admit they outsource to Tier 2.
• 🌍 Exporters won’t reveal source farms — they fear being cut out.
• 🏢 Traders obscure price breakdowns to protect margins.
• 🛠 Manufacturers worry about audits, and buyers asking questions.

If you ask for documents, many will send you PDFs.
If you ask where the data came from, the conversation ends.

💰 Opacity is profitable.
Hiding intermediaries = higher markup.
Faking compliance = cheaper business.

🧨 Transparency is risky.
Disclosing weak links = losing contracts or getting fined.

🤷 There’s no shared map.
So bad actors win by default —
because they can pretend better than they can prove.

If we want ethical trade,
then we need to make truth a competitive advantage.

If the playing field rewards hiding, then honesty becomes a disadvantage.

So a real solution doesn’t just collect data —
it shifts the incentives in the system.

It must:

• 🧭 Make visibility a strength, not a risk
• 🛠 Work without forcing new tools on small players
• 🧠 Reason across missing or fuzzy data
• 🔍 Show how companies connect — not just who they are
• 🧾 Make it easier to prove the truth than to fake it

In short:

🧱 It’s not about reporting tools.
It’s about building a new type of evidence.

🧠 Understands context and gaps
“Sourced through ChinaCorp” ≠ “Sourced from ChinaCorp”

🔗 Reveals hidden structures
Not just documents — relationships

🧩 Integrates finance, movement, and meaning
Supply chains are physical, economic, and narrative systems

✅ Gives good actors a chance to win
If you can prove it, you can compete

SupplyLens is an open-source tool that helps you map the reality of supply chains —
even when the data is fragmented, hidden, or manipulated.

It doesn’t just store documents.
It reasons across them.

It combines:

• 📥 Smart data ingestion (from PDFs, websites, leaks, forms)
• 🧠 LLM-based reasoning to extract relationships and meaning
• 🔗 Graph database structure to reflect how supply chains actually behave
• 🔍 Semantic search and risk detection, even when things aren’t said explicitly

And it works with the world as it is —
Not as we wish it were.

📊 Built on PostgreSQL + Apache AGE + pgvector
One schema, no microservice chaos

💬 Natural language interface
“Where is EasyFiets’ lithium most likely from?”

🧠 AI + graph logic = supply chain insight
It thinks through the chain — not just shows it

🛠 Designed for:
- Researchers
- Universities
- ESG teams
- Investigative journalists
- Supply chain innovators

SupplyLens processes messy, fragmented data — and makes meaning from it.

It takes in:

• 📄 PDFs, websites, spreadsheets, or scans
  
• 🔍 NGO reports, invoices, product catalogs
  
• 🗂 Structured or unstructured documents

Then it:

• 🧠 Extracts entities, relationships, and sources
  
• 🔗 Builds a graph of companies, flows, risks, and gaps
  
• 🧾 Annotates each node with metadata, ESG claims, and evidence
  
• 🧭 Uses AI to infer missing links or weak spots
  
• 💬 Responds to natural language questions about the chain

You don’t need 100’s of records. You don’t need a data team.
You just need a question — and SupplyLens will trace the answer.

📜 EU Laws Require This

• CSRD: ~50,000 EU-based firms must report on full ESG impacts
  → Corporate Sustainability Reporting Directive, 2023
  normative.io › CSRD Explained

• CSDDD (CS3D): Human rights & environmental due diligence must go beyond Tier 1
  → EU Corporate Sustainability Due Diligence Directive, April 2024
  morganlewis.com › CSDDD Adopted

📊 Research-Backed

• LLM + Knowledge Graphs improve traceability without direct supplier input
  → “Supply Chain Mapping with LLMs and Graph AI” (2024 preprint)
  arxiv.org › 2408.07705v1

🗂️ Data-Ready

• WikiRate: 7M+ ESG datapoints from 145k+ companies
  wikirate.org

• Open Supply Hub: 100k+ mapped facilities & contributors
  opensupplyhub.org

Christiaan Verhoef
Vision & Business
LinkedIn

Torsten Raudssus
AI Engineering
LinkedIn

Prof. Michiel Steeman
Strategic Support
LinkedIn