Aligning event logs and process models for multi-perspective conformance checking : an approach based on integer linear programming

Modern organizations have invested in collections of descriptive and/or normative process models, but these rarely describe the actual processes adequately. Therefore, a variety of techniques for conformance checking have been proposed to pinpoint discrepancies between modeled and observed behavior. However, these techniques typically focus on the control-flow and abstract from data, resources and time. This paper describes an approach that aligns event log and model while taking all perspectives into account (i.e., also data, time and resources). This way it is possible to quantify conformance and analyze differences between model and reality. The approach was implemented using ProM and has been evaluated using both synthetic event logs and a real-life case study