Linking the Chain: Linked Data and Knowledge Graphs in commodity chains

Globalization increases the interconnection of markets, resulting in longer and more internationalized supply chains and thus higher complexity, making tracing products and their components challenging. For proper risk assessment, products and theircomponents need to be reliably identifiable and traceable. Due to the number of partners and methods involved, collected data will occur in different formats, both as structured and unstructured data, making it difficult to utilize the full potential of the

data available. Linked Data can help integrate data from different sources, interlinking
and storing relations between data points and therefore, simplify the analysis of this
complex and interconnected information. The developments in large language models
have made it possible to at least partially automate the task of transforming data into
Linked Data. We will show the process of generating Linked Data with an example data
set, including triple generation and matching of terms to existing ontologies. Especially
commodity chains can benefit from Linked Data in Knowledge Graphs as information
across different stages in the supply chain can be related, seamlessly processed and
analyzed. Knowledge Graphs allow discovering hidden relationships and therefore help
getting a holistic view on commodity chains, avoiding data silos. This bears the
possibility to consider additional factors for risk assessment and to act proactively, for
example when it comes to product recall scenarios. To conclude, Linked Data and
Knowledge Graphs can facilitate getting a holistic view on complex data and by doing so can help to identify risk factors along the commodity chain.