Mining process flowcharts from business data: An evolutionary approach

Abstract

The practice of process mining concerns the reconstruction of complete process flow diagrams from data logs. In an increasingly complex business environment, there is a need for managers to understand the processes they already have in place. Process mining facilitates the automatic reconstruction of a flowchart description based on a set of execution traces. The process mining approach featured in this paper utilises a Genetic Programming approach to process mining using a graph representation. The results from a number of experiments are detailed in this paper and point to its potential as a practical tool for use in industry. The principles investigated in this approach form a core component of future research by the authors in the area of process disparity identification. © Springer-Verlag Berlin Heidelberg 2010.