Bridging the Gap: How Process Mining Practitioners and Researchers Address Data Quality Issues

Abstract

Process mining integrates process science and data science to analyze process workflows using event logs. Initially an academic discipline, it has seen rapid adoption in industry, often combined with machine learning and automation. This study explores how researchers and practitioners approach data quality issues found in event logs and how they apply preprocessing techniques to minimize or solve said issues. Results show that practitioners often undervalue data quality challenges and rely on basic methods, likely due to limited experience and dependence on commercial tools like Celonis. On the other hand, researchers prioritize diverse and advanced preprocessing techniques and view data quality issues as critical in process mining projects. Respondents with dual roles demonstrate specific expertise, addressing diverse challenges with data quality issues and applying more complex preprocessing techniques. The study emphasizes the need for collaboration between academia and industry, integrating process mining into education, and enhancing tool capabilities. These steps can bridge knowledge gaps, promote best practices, and advance research and practical application in process mining.

Article history: Received (February 11, 2025); Revised (February 27, 2026); Accepted (March 19, 2026); Published online (May 9, 2026)