Data Quality in Process Mining

Abstract

To cope with challenges such as tightening budgets and increased care needs, healthcare organizations are becoming increasingly aware of the need to understand their processes in order to improve them. In this respect, process mining has the unique potential to retrieve process-related insights from process execution data. Despite the wide range of algorithms that have been developed over the past decade, the reliability of process mining outcomes ultimately depends on the quality of the input data. Consistent with the notion of “Garbage In, Garbage Out”, applying process mining algorithms to low quality data can lead to counter-intuitive or even misleading decisions. Real-life healthcare event logs typically suffer from a multitude of data quality issues such as missing events, incorrect timestamps and incorrect resource information. Against this background, this chapter provides an introduction to data quality in the process mining field. Three key topics are discussed: (1) data quality taxonomies, i.e. frameworks outlining potential data quality issues, (2) data quality assessment, i.e. the identification of data quality issues, and (3) data cleaning, i.e. efforts towards alleviating data quality issues which are present in an event log.