A Multi-perspective Online Conformance Checking Technique

Abstract

Process mining is a field of research which deals with the extraction of information about processes, from their execution event logs. Conformance checking, a main type of process mining, confronts an existing process model with an event log of the same process to detect, locate and explain the deviations between them. The vast majority of the conformance checking techniques focus only on the control-flow perspective (i.e., the ordering of the performed activities) and can be used only in an off-line setting, even though deviations can occur anytime and in other perspectives (e.g., time, resource or data) as well. In this work, a Multi-perspective Online Conformance Checking (MOCC) technique is introduced to support real-time monitoring of event data from various angles. MOCC can detect deviations in real-time and in any perspective, so recovery actions can be performed (in some cases) even before the execution of a process instance is completed.