Process Mining
What is Process Mining?
Process mining is an exciting scientific discipline which combines interesting research challenges with a high practical value. More and more events are recorded by a wide variety of systems (cf. internet of things, Industry 4.0, social media, mobile devices, web services, etc.). The spectacular growth of event data provides many opportunities for automated process discovery based on facts. Moreover, event logs can be replayed on process models to check conformance and analyze bottlenecks. The uptake of process mining is reflected by the growing number of commercial process mining tools available today. There are over 25 commercial products supporting process mining (Celonis, Disco, Minit, myInvenio, ProcessGold, QPR, etc.). All of these products support process discovery and can be used to improve compliance and performance problems. This way the fruits of process mining research are transferred to real-life applications.
Process mining bridges the gap between traditional model-based process analysis (e.g., simulation and other business process management techniques) and data-centric analysis techniques such as machine learning and data mining. Process mining seeks the confrontation between event data (i.e., observed behavior) and process models (hand-made or discovered automatically). This technology has become available only recently, but is mature enough to be applied to care processes of any type and of any complexity. The process-mining spectrum is broad and includes techniques for process discovery, conformance checking, prediction, and bottleneck analysis. Traditional data-mining approaches are not process-centric. Input for data mining is typically a set of records and the output is a decision tree, a collection of clusters, or frequent patterns. Process mining starts from events and the output is related to an end-to-end process model. Data mining tools can be used to support particular decisions in a larger process. However, they cannot be used for process discovery, conformance checking, and other forms of process analysis. Therefore, process mining is different from mainstream data science and process science techniques. However, it is also integrative in nature. In short, it already widely applicable but also provides many new research challenges.