Securing the supply chain

A French energy group manages over 8,500 suppliers worldwide, from critical raw materials to highly specialized maintenance services. How do you anticipate a failure before it happens?

Request a demo

Anticipating supplier risk

Indirect dependencies stay invisible

The purchasing department was facing geopolitical instability, rising energy costs and new CSR constraints (duty of vigilance law, CSRD). Teams regularly noticed unforeseen shortages on strategic categories: spare parts for power plant maintenance plans, raw materials, specialized expertise.

Traditional e-procurement tools couldn't reveal tier-2 and tier-3 dependencies, nor simulate the impact of a failure, let alone anticipate a geopolitical risk affecting an extended supplier.

WHAT E-PROCUREMENT DOESN'T SEE

44%

of the components on a strategic production line depended on a single supplier, based in a geopolitically unstable region.

The fragility point no one had quantified

By modeling the whole supply network as a graph, the mapping made critical dependencies visible, including at indirect subcontractors. This figure didn't come out of a months-long audit: it was already in the data, it just lacked the structure to read it.

Six entities, six relationships

THE PURCHASING ONTOLOGY

A project can also DEPEND_ON a supplier, directly or indirectly.

The supply network as a graph

Supplier : country, tier 1/2/3, certification, ESG score, criticality. Component : category, criticality, unit price, carbon footprint. Project : location, sector, critical dependencies.

Order : date, amount, status (in progress, delivered, blocked). Region : geopolitical risk, energy risk, regulatory compliance. Audit : date, result, ESG and compliance anomalies.

The analysis path, in four steps

01 Dependency analysis Four projects depend on the same critical supplier F0093. The concentration shows up as a shape in the graph, not a line in a table.

02 Impact analysis F0093 supplies components M0162 and M0104: you know immediately what goes down with it, and on which projects.

03 Finding alternatives The same components can be supplied by low- or moderate-criticality suppliers, filtered directly in the view.

04 Compliance check Along the way, you check that these alternatives were audited recently — otherwise, you request the audit before committing.

ALTERNATIVES TO F0093 — M0162 & M0104

SUPPLIERCRITICALITYLAST AUDIT F0093 critical current panel F0148 low 03/2026 F0206 moderate 11/2025 F0311 low audit to request

The same screen filters criticality and checks how recent the audits are.

Simulate before deciding

Buyers test different scenarios: a key supplier shutting down, raw material price hikes, and measure the cascading effects on ongoing projects.

These analyses made it possible to diversify the supplier panel, strengthen strategic partnerships, and integrate ESG criteria into tenders to reduce regulatory exposure.

The results

− 34 %

fewer critical-risk suppliers, thanks to panel diversification.

95 %

time saved compared to the initially deployed solutions.

3 mois

of extra visibility on supply-disruption scenarios.

Easier regulatory reporting

CSR and "duty of vigilance" reports are produced from the same structure: presenting the network and the audits is enough.

Stronger negotiating position

Thanks to fine-grained total-cost-of-ownership analysis, supplier by supplier and component by component.

Purchasing departments move from a reactive, fragmented logic to a proactive, systemic approach: modeling not just the data, but its relationships and how each piece affects the others.

Where is your 44% concentration?

We map a real purchasing category (tiers 1 through 3 included) and simulate the failure of the supplier that worries you most.

Let's talk

Newsletter

Subscribe to our newsletter

Insight on knowledge graphs, straight to your inbox.

FROM SCATTERED DATA TO RELIABLE DECISIONS

A team that comes from the field

Shedding light on graph technology and its applications.

One platform, many ways to use it