Knowledge Graphs and Reasoning to Capture Product Complexity in Factory Automation

The talk reports about a deployed semantic solution for a complex combinatorial product configuration problem at Festo, a global leader in factory automation. It replaces a RDBS solution whose conceptual and practical capabilities failed to meet the productive requirements in terms of performance, maintenance as well as flexibility. The semantic replacement was developed in close cooperation with derivo GmbH and makes extensive use of Knowledge Graphs, OWL RL and SWRL reasoning, and SPARQL querying.

The main part of the talk will provide insights to the following aspects:
• staged Knowledge Graph generation from legacy data source with R2RML
• data enrichment by OWL RL inference
• computation of product compatibility with OWL RL and SWRL reasoning
• modelling of technical and business-related constraints on product data
• ad-hoc querying a multi-million combinatorial product composition graph

Speakers: