Event abstraction in process mining: literature review and taxonomy

Sebastiaan J Van Zelst,Agnes Koschmider,Massimiliano De Leoni,Felix Mannhardt

doi:10.1007/s41066-020-00226-2

Abstract

The execution of processes in companies generates traces of event data, stored in the underlying information system(s), capturing the actual execution of the process. Analyzing event data, i.e., the focus of process mining, yields a detailed understanding of the process, e.g., we are able to discover the control flow of the process and detect compliance and performance issues. Most process mining techniques assume that the event data are of the same and/or appropriate level of granularity. However, in practice, the data are extracted from different systems, e.g., systems for customer relationship management, Enterprise Resource Planning, etc., record the events at different granularity levels. Hence, pre-processing techniques that allow us to abstract event data into the right level of granularity are vital for the successful application of process mining. In this paper, we present a literature study, in which we assess the state-of-the-art in the application of such event abstraction techniques in the field of process mining. The survey is accompanied by a taxonomy of the existing approaches, which we exploit to highlight interesting novel directions.

Highlights

In modern organizations, the execution of business processes is often supported by different information systems
Mixed-granular event data hampers the applicability of process mining, i.e., discovered models tend to be too complex and conceal the true business-level process logic
We present the results of a systematic literature review, covering work that abstracts events into a higher level of granularity

Summary

Introduction

The execution of business processes is often supported by different information systems. Process mining provides several techniques to extract actionable knowledge and insights of a process, on the basis of historical execution data (van der Aalst 2016). Within the realm of process mining, process discovery algorithms are able to translate the captured event data into a process model, in a (semi)automated fashion. Conformance checking algorithms allow us to compute whether or not the execution of the process, as recorded in the event data, is in line with a reference model. Several techniques exist that allow us to compute insights in the performance of the process, perform root-cause analyses, correlate behavior with different KPIs, improve processes and its models, etc., see, e.g., van der Aalst (2016)

Methods

Findings

Discussion

Conclusion