As companies push for digital transformation, they are accumulating more data about their routines, both structured and unstructured. Process mining lets organizations convert their operational event data into meaningful insights. Such insights can help them identify issues and enhance their processes. Various software solutions for process mining have been developed, both for industrial and academic purposes, but most of these solutions do not support the creation and execution of analytical workflows.
To that end, the Process Mining group at the Fraunhofer Institute for Applied Information Technology (FIT) has developed an extension in collaboration with KNIME: PM4KNIME implements state-of-the-art techniques for discovering process models, checking conformance between event data and process models, and composable visualizations (which support data apps). It leverages algorithms of ProM, one of the most powerful academic process mining tools comprising hundreds of plugins. PM4KNIME effectively facilitates the user-friendly creation of workflows combining advanced process mining with the built-in data science and analytics capabilities of KNIME.
Algorithms in PM4KNIME have been adapted to work natively with KNIME tables instead of dedicated XES logs. This allows for the use of many functionalities of the KNIME Analytics Platform, such as data filtering, processing, and even handling unstructured data. Applying process mining algorithms directly to KNIME tables further increases the performance because KNIME’s powerful caching strategies ensure high scalability when processing large data tables.
About the Process Mining group at FIT
The Process Mining group at the Fraunhofer Institute for Applied Information Technology runs various process mining initiatives, ranging from development to application. They actively contribute to cutting-edge research in the field. The group works on multiple projects, including large research endeavors, applications of process mining at industrial partners, and development of various process mining frameworks and software solutions.
Explore More Resources
-
Install PM4KNIME from Community Hub and explore examples
-
Read the PM4KNIME paper presented at ICPM 2022
-
Observe development of PM4KNIME on Github
-
Join the discussion on the KNIME Forum