Handling Inconsistencies in Business Process Modelling

Project description

Business rules and business process models are widely acknowledged as a useful means for the organization of businesses. Both these artifacts are usually created and maintained by human modelers. Here, a potential problem is that of inconsistency, which can result from an incremental and collaborative modelling process. For example, Batoulis and Weske (2017) report on a recent case-study with a large insurance company, where they found that 27% of business rules were erroneous. In particular, in the case of digitization, with processes being accessed across socio-technical systems, multi-cultural or multi-jurisdictional boundaries, business process management (BPM) and business rules management (BRM) becomes even more challenging. As business rules are the de facto object of study for compliance management, where rules are utilized to govern compliant processes, such inconsistencies are also a problem for compliance management, as inconsistent rule bases cannot be used to verify compliance correctly, which thus impedes company efforts in compliance management. Here, methods are needed to detect inconsistencies and present companies with a careful analysis to support inconsistency resolution. Also, different users may model aspects of a scenario differently and joining their partial models may yield unsatisfiable processes, meaning an adaptation of the model is called for. In these scenarios, methods are needed that analyse the contradiction and present the users with a analysis so they may be able to re-design the model in an appropriate manner or trigger (semi)-automatic methods to repair the model. This calls for both, a qualitative analysis of the reasons of contradictions (such as pin-pointing the exact cause of problems) and a quantitative analysis of the severity of problems. We therefore aim to investigate the detection, analysis, and resolution of inconsistencies in business rules and processes. In the envisioned project, we aim to address the challenge of dealing with erroneous process models and business rules by employing methods from inconsistency measurement, a research topic within the area of knowledge representation and reasoning. Our aim is to develop measures that can assess the severity of inconsistencies occurring within business rules or between process models and business rules, and present respective results in a way comprehensible for stakeholders in BPM and BRM. These quantitative insights can be used for inconsistency detection and ranking and support subsequent inconsistency resolution in the context of business process improvement. Also, we will investigate means for (semi-) automated inconsistency resolution, as well as guiding modellers in resolving errors in an incremental-manner.


Project leader

People

Publications

  • Carl Corea, Patrick Delfmann. Quasi-Inconsistency in Declarative Process Models. In Proceedings des Business Process Management Forum (BPM2019). Wien, 2019.
  • Carl Corea, Matthias Deisen, Patrick Delfmann. Resolving Inconsistencies in Declarative Process Models based on Culpability Measurement. In Proceedings der 14. Internationalen Tagung der Wirtschaftsinformatik (WI2019). Siegburg, 2019. Gewinner des Best Paper Awards.
  • Sabine Nagel, Carl Corea, Patrick Delfmann. Effects of Quantitative Measures on Understanding Inconsistencies in Business Rules. In Proceedings of the 52nd Hawaii International Conference on System Sciences (HICSS52). Hawaii, 2019
  • Carl Corea, Patrick Delfmann. Supporting Business Rule Management with Inconsistency Analysis. In Proceedings of the Industrial Track at BPM 2018 co-located with 16th International Conference on Business Process Management (BPM 2018). Sydney, 2018
  • Carl Corea, Patrick Delfmann. A Tool to Monitor Consistent Decision-Making in Business Process Execution. In Proceedings of the Demonstration Track at BPM 2018 co-located with 16th International Conference on Business Process Management (BPM2018). Sydney, 2018



Last updated 02.07.2019, Matthias Thimm | Terms