Generating event logs for high-level process models

Abstract

Business Process Model and Notation (BPMN) is a de-facto standard for practitioners working in the Business Process Management (BPM) field. The BPMN standard [1] offers high-level modeling constructs, such as subprocesses, events, data and message flows, lanes, and is widely used to model processes in various domains. Recently several BPMN-based process mining techniques [2, 3, 4] were introduced. These techniques allow representing processes, discovered from the event logs of process-aware information systems, in a convenient way, using the BPMN standard. To test these mining approaches an appropriate tool for the generation of event logs from BPMN models is needed. In this work we suggest such a tool. We propose a formal token-based executable BPMN semantics, which takes into account BPMN 2.0 with its expressive constructs. The developed tool is based on these semantics and allows simulation of hierarchical process models (including models with cancellations), models with data flows and pools, and models interacting through message flows. To manage the control flow, script-based gateways and choice preferences are implemented as well. The proposed simulation technique was implemented on top of existing plug-ins for ProM (Process Mining Framework) [5], and was verified on models created by practitioners from various domains.