How AI and Knowledge Graphs Strengthen Supply Chain Resilience

In our complex world, even the most sophisticated supply networks have become highly fragile systems. eccenca and the research consortium of the CoyPu project addressed this challenge with an Artificial Intelligence platform for integrating, structuring, networking, analyzing and evaluating heterogeneous data from economic value networks as well as the industry environment and social context. The aim: to strengthen resilience in the supply chain.
Resilient supply chains require context, not just data
For a long time, supply chains were a silent constant, running in the background and functioning reliably - until they stopped doing so. In recent years, it has become clear how vulnerable even the most sophisticated supply networks can be. Well-known causes include global crises, material shortages, and political instability. However, the problem often lies much closer to home - in the lack of visibility and structure of supply chain data. Most of the supply chains are data-blind.
The reality: supply chains are data-blind
Many companies hold large amounts of information – about suppliers, material flows, production locations and risk indicators. But this data is:
- spread across departments, systems, and locations,
- often isolated, incomplete, or outdated,
- not interoperable or lacking consistent standards,
- difficult to understand for both human and machine users.
The result: there is no foundation for reliable decision-making. Resilience becomes a reactive damage limitation rather than a strategic capability.
The CoyPu project: when artificial intelligence strengthens ecosystems
With the CoyPu (Cognitive Economy Intelligence Platform for the Resilience of Economic Ecosystems) research project - funded by the German Federal Ministry for Economic Affairs and Energy (BMWi) as part of the ‘Artificial intelligence as a driver for economically relevant ecosystems’ initiative - eccenca and the CoyPu research partners have pursued one goal: Making supply chains more resilient through contextualised data analysis.
To this end, research was conducted into solutions that address structural weaknesses in the information base. The focus was not only on collecting data but also on making it comprehensible - for humans and machines alike.
The project brought together a strong consortium of academic, research, and industry partners, including Datev eG, Implisense GmbH, Siemens AG, Infineon Technologies AG, Institut für Angewandte Informatik e.V. (INFAI), Hamburger Informatik Technologie-Center e.V. (HITeC), Forschungszentrum L3S, Leibniz Universität Hannover, Leibniz Informationszentrum Technik und Naturwissenschaften (TIB), Deutsches Institut für Wirtschaftsforschung (DIW), Selbstregulierung Informationswirtschaft e.V., and others.
What has been developed?
The project partners developed an intelligent AI-powered platform that integrates, semantically structures, and analyzes vast and heterogeneous economic data—using hybrid approaches combining knowledge graphs and machine learning — to support real-time decision-making during crises. It features capabilities such as risk detection, trend analysis, network modeling, forecasting, and interactive dashboards to enhance resilience across companies, industries, and regions.
The key innovations contributed by eccenca include:
- Semantic linking of isolated data sources: Data from ERP, SCM, Excel, web portals, and partner networks is brought together on a shared knowledge base [SK1] - automatically and comprehensibly.
- Visualisation of value creation networks in the knowledge graph: Supply chains are not only mapped but also depicted as living structures. Dependencies, risks, and alternatives can thus be identified more quickly.
- Graphical test interface for data quality: A visual tool shows directly where data gaps or conflicts exist – making data maintainance an easy to do for both technical and non-technical users.
- Benchmark procedure for mapping tools: A new procedure for evaluating automated data links has been developed and prototypically integrated into the knowledge graph platform eccenca Corporate Memory. This allows users to systematically assess the quality and efficiency of data mappings, compare different tools, and make informed decisions about their integration workflows

From research to application: eccenca Corporate Memory
The findings and components from CoyPu are incorporated into our eccenca Corporate Memory platform. It enables companies to
- semantically integrate heterogeneous data sources,
- make knowledge about supply chains contextually accessible,
- analyse risks and vulnerabilities based on data,
- and to make data-driven decisions in a more resilient way.
In other words, eccenca corporate Memory with all its features allows you to enrich your data with context, so it becomes actionable - for people, for machines, for future scenarios.
Why is this important?
Resilience is not just a buzzword. It determines whether companies remain able to deliver, serve markets, or develop strategically. However, resilience requires visibility. It needs data, but above all, it needs context. This is the only way to turn information into action.
Learn more about
- the CoyPu project
- eccenca as a solution for digital supply networks
- the platform eccenca Corporate Memory
- eccenca Research - Projects & Innovations