Model-Driven Context Configuration in Business Process Management Systems: An Approach Based on Knowledge Graphs

Abstract

Business Process Management Systems (BPMSs) are inherently model-driven, relying on machine-readable process repositories that are typically standards-based. However, a requirement for semantic agility is emerging as knowledge-driven applications become less blueprint-oriented and more context-aware. The integration of process knowledge with contextual data can be subjected to this agility requirement – i.e., having the process modelling environment customised in terms of (expanding) its knowledge space and in terms of model-data interoperability. Such customisations may capture any of the enterprise perspectives proposed by the Zachman Framework (among which the How, Who and Where facets are in our particular focus) towards the benefit of establishing a hybrid knowledge-data fabric underlying flexible, context-driven BPMSs. This paper presents a project-based technical solution, based on the interplay of semantic technology and agile modelling methods, for setting up a hybrid knowledge base derived from several heterogeneous sources: diagrammatic models, semantically lifted legacy data and open geospatial data, with reasoning rules on top of this conglomerate. Together, these sources cover the How, Who and Where facets of the Zachman Framework concepts in a Knowledge Graph that drives the front-end Task Management panel of a BPMS. The proposal advocates complementarity and integration of paradigms that rarely converge – i.e., knowledge representation, open data and process-aware information systems.