Ontology-Based Multi-perspective Process Mining in Laboratories: A Case Study

In real-world process mining scenarios, event logs are extracted from relational databases. The de-facto standard for event log exchange, XES (eXtensible Event Stream), has the limitation that it is restricted to a single case notion and hence viewpoint. Although recent trends towards object-centric process mining propose new standards, they are not yet widely adopted by the tool industry. In this paper, we show how to enable multi-perspective process mining based on the established XES standard by means of a case study from the laboratory domain. Inspired by an existing approach, we use semantic technology to enable the mapping between a domain ontology and a relational database. This allows to dynamically switch the perspective with every new query. In addition, we employ the semantic extension of XES to enrich the resulting event log with semantic information from the domain ontology. Our prototype has been evaluated on real-world data from a German laboratory and therefore specifically addresses the challenge on how to process protected health information.