Discovering Local Concurrency Relations in Business Process Event Logs

Detecting concurrency relations between events is a fundamental primitive in process mining. Existing approaches to this problem identify concurrency relations between pairs of event types under a global interpretation. If two event types are found to be concurrent, every occurrence of one event type is deemed to be concurrent to one occurrence of the other. In practice, this assumption not always holds. This paper proposes a finer-grained approach, whereby two event types may be in a concurrency relation relative to one state of the process, but not relative to other states, i.e. the concurrency relation holds locally. Experimental results both with artificial and real-life logs show that the proposed local concurrency detection approach improves the accuracy of existing concurrency detection techniques.