A Case-Based Approach for Workflow Flexibility by Deviation.

This paper presents a case-based approach for workflow flexibility by deviation. In previous work, a constraint-based workflow model and engine have been developed that allow for flexible deviations from predefined workflow models during run-time. When encountering deviations, domain-independent strategies can be applied for a resolution in order to regain support for the process participant. To improve this deviation handling, a case-based approach is presented that integrates experiential knowledge by exploiting previously terminated workflows as cases. Similar cases are retrieved through a time-series based similarity measure and reused through null adaptation. The experimental evaluation showed an improvement of the defined utility value concerning the computed work items, when comparing the constraint-based workflow engine and the case-based deviation management.