Hierarchical Business Process Discovery: Identifying Sub-processes Using Lifecycle Information

Abstract

This paper aims to introduce a novel approach to discover hierarchical business process models from event logs with lifecycle information. To handle noise and infrequent behavior, we introduce the notion of nesting ratio to quantify the probability of nesting. All proposed approaches have been implemented in the open-source process mining toolkit ProM. The proposed approach is compared to existing process discovery techniques using both synthetic and real-life lifecycle event logs, and finally we show that our approach outperforms exiting approaches to discover hierarchical processes.