Similarity Measurement for Ontology Based Product Configurations

Kurzfassung

Many technical devices are complex systems with a wide range of configuration options. Such options can be selected, but not independently and with restrictions and implications. The matching between customers requirements to existing configurations that a supplier can build with sensible effort and technical limitations is currently mostly done manually based on experience. In this paper, we present an approach to automate this matching, based on ontologies and (tailored) similarity measures. We explain an implementation that works on a reduced parameter set for star trackers, based on the example of ASTRO APS® from Jena­Optronik GmbH. This system is able to automatically suggest the best fitting and buildable configuration for any requirement configuration and we point out possible ways to extend it in the future.