Project: Multi-Facet BPM

Team: M. Sc. Tom Thaler, PD Dr. Peter Fettke

Research institution: Institute for Information Systems (IWi) at the German Research Center for Artificial Intelligence (DFKI) and Saarland University

Abstract: Business Process Management is a manifold field in the area of information systems research. It consists of the phases strategy, design, implementation, execution, monitoring & controlling and improvement of business processes, whereby each of these phases has its own challenges. These challenge range from methodological ones (e. g. how to design process models and are there best practices modeling them) to technical ones (e. g. in terms of software architecture paradigms or the complexity of particular algorithms). Also combinations like the technical support for process design or the management and handling of large business process repositories are important research topics.

As the aspect of Big Data plays an increasing role in the context of business process management in general and process mining in particular, the project “Multi-Facet BPM” addresses the following three scenarios: (1) Study the state-explosion problem in real applications, (2) trace extractions of business process models and (3) process mining on Big Data.

Similarity matrix as heatmap Figure 1: Similarity matrix as heatmap

One of the core fields in process mining is conformance checking, whereby a node matching between models is needed. One part of our work covers the calculation of those matchings with different techniques on a large process model repository, which contains 2.300 single models with about 20.000 events, 20.000 functions and 10.000 connectors. Thereby the similarity between all nodes of a process model to all nodes of all other models must be calculated, which leads to the need of a high performance it infrastructure. These node matchings are also the basis for the calculation of process model similarities, which are visualized in Figure 1. The green blocks show clusters of process models (rows and columns) with high similarity.

Last modified 5 years ago Last modified on Feb 5, 2014 2:50:12 PM

Attachments (1)

Download all attachments as: .zip