Data-driven process control and exception handling in process management systems

Business processes are often characterized by high variability and dynamics, which cannot be always captured in contemporary process management systems (PMS). Adaptive PMS have emerged in recent years, but do not completely solve this problem. In particular, users are not adequately supported in dealing with real-world exceptions. Exception handling usually requires manual interactions and necessary process adaptations have to be defined at the control flow level. Altogether, only experienced users are able to cope with these tasks. As an alternative, changes on process data (elements) can be more easily accomplished, and a more data-driven view on (adaptive) PMS can help to bridge the gap between real-world processes and computerized ones. In this paper we present an approach for data-driven process control allowing for the automated expansion and adaptation of task nets during runtime. By integrating and exploiting context information this approach further enables automated exception handling at a high level and in a user-friendly way. Altogether, the presented work provides an added value to current adaptive PMS.