Graph Models for Knowledge, Regulations, Rules and Processes

In this DIREC TALK Thomas Hildebrandt presents how graph models can be used for representation of machine-readable regulations, rules and distributed processes in a flexible and maintainable way supporting both human understanding and automated execution.

Computer Science deals with the theory and methods for designing, analyzing and engineering systems of data and processes used by and impacting people and the society in which they are embedded. An important ingredient is the development of formal languages and structures for describing data and processes that can at the same time capture the complexity of the problem domain and be subject for analysis and execution by computers.

As the technology and use of computers has evolved and changed over time, a plethora of different languages and structures have been introduced.

Mirroring the evolution from centralized computer systems used mainly for business processes and research to ubiquitous, distributed systems handling processes spanning both our professional and private lives, a key challenge has become the design, analysis and management of distributed and frequently changing structures of data and processes and the regulations and rules they are supposed to follow.

Concretely, Thomas Hildebrandt will present the theory and tools of Dynamic Condition Response (DCR) Graph and give concrete examples of the modelling of legal regulations, rules and processes. The theory and tools are the result of more than 15 years of research and development jointly with industry and public organizations culminating in the establishment of the company in 2018 providing industrial strength tools for design, analysis and execution of decision and process models, which has so far been embedded in the widely used WorkZone enterprise information management system from KMD/NEC as well as open source case management systems used in municipalities in Denmark.

The talk will conclude with directions for current and future research, including the relation between explainable AI and DCR graphs and the award winning process mining based on DCR Graphs and how to represent more general knowledge of organisations which is currently peaking several of Gartner’s hype curves (e.g. Emergent Technologies and Government Technologies and AI) under the terms like Human-centered AI, knowledge graphs, decision intelligence and Digital twins of Government.





Thomas Hildebrandt is professor in software engineering and Head of Software, Data, People & Society research section at University of Copenhagen. With a background in formal process models he has in more than 10 years been leading inter-disciplinary research and innovation projects with focus on methods and technologies for developing reliable and flexible software systems suited for the people who use them, including digitalisation of law, workflows and business processes information systems.

The research carried out by Thomas has lead to the development of the process technology Dynamic Condition Response (DCR) Graphs in collaboration with the company Exformatics. The technology has users all over the world and is available as a service at DCR is now owned by the company DCR Solutions and is used in Denmark to support flexible case management within the KMD WorkZone case management system, which is used 65% of the employees in the Danish state, including administrative workers at several universities.