Process Mining: Applied Discipline and Software Implementations

Abstract

Background. A precise picture of how business processes (in the interpretation by Andrea Burattin) are performed in real-life is vitally important for an organization because it shows actual situation revealing gaps and bottlenecks. Process mining is a discipline with the purpose to research processes using as the input so-called event data (or event logs) which in essence is a digital footprint left in IT systems as the result of business processes execution. Objective. The goal of the study is to overview current state of process mining and find actual scientific and practice tasks in this field as well as justify and formalize requirements to the information technologies with the purpose to implement the found set of process mining applied tasks. Methods. The method used by the authors to prepare current overview consisted of the following steps: (a) analysis of information sources; (b) finding and formalization of actual scientific as well as practice tasks; (c) description of the requirements to the information technologies with the purpose to implement the found set of actual tasks. Results. It has been found out that process mining as an applied discipline has been actively developed for 20 years; significant contribution to creating the scientific basis of process mining has been done in Eindhoven University of Technology (The Netherlands) under direction of professor Wil M.P. van der Aalst. It also has been found actual scientific and practice tasks of process mining: event data preparation, dealing with concept drift, operational support, event data streams processing, handling big event data, improving process mining tools usability for the end users. It has been formalized requirements and specified quality attributes for the information technologies with the purpose to implement the found actual tasks. Architecture of the information technologies has been proposed by the authors. Conclusions. Currently the theoretical core of process mining has mainly been developed and quite structured. However, despite of the fact that mathematical methods and software tools have been successfully used in practice for a few years, the request for the intellectual business process analysis has not been fulfilled yet. The authors have found out that relevant information technologies should supply such functions as handling big event logs, dealing with event data streams as well as operational support of business processes which are at the execution stage.

Keywords

• Process mining
• Information technology
• Business process management
• BPM
• Event data
• Event logs
• XES, ProM
• Disco
• Celonis