IEEE CIS Task Force on Process Mining

Trace: 2010_09_29_bpm


Minutes of the meeting of the Task Force at BPM 2010

These are the minutes of the IEEE Task Force on Process Mining Meeting at BPM 2010 September 15th 2010, Stevens Institute of Technology, Hoboken NJ.

Although the meeting was running in parallel with other sessions (including presentations of task force members and process mining related demo's), it was well visited. About 20 people attended.

Wil van der Aalst introduced some basic facts about the task force which was established in 2009. Alessandro Sperduti played a key role in the creation of the task force and Wil van der Aalst has been chairing the task force since its inception. The task force groups various organizations: software vendors (Pallas Athena, IDS Scheer/Software AG, Futura Process Intelligence, HP, IBM, Infosys, Fluxicon, Businesscape, Iontas, Fujitsu, Business Process Mining), consultancy-related organizations (some of the above and ProcessGold, Business Process Trends, Gartner, Deloitte), and universities (TU/e, University of Padua, University of Catalunya, New Mexico State University, Technical University of Lisbon, University of Calabria, Penn State University, University of Bari, Humboldt-Universität, Stevens Institute of Technology, University of Haifa, Seoul National University of Technology, Cranfield University, K.U. Leuven, Tsinghua University, Innsbruck University). The goal of the Task Force is to promote the research, development, education and understanding of process mining.

After this introduction, the various participants of the meeting introduced themselves. Subsequently, the discussion focused on the promotion of the topic. Wil van der Aalst shows a just list of initiatives (see slides) that have taken place. Also the upcoming special session on process mining at IEEE Symposium on Computational Intelligence and Data Mining ( in Paris (April 2011) was mentioned and task force members are encouraged to submit papers. However, it seems that one of the key elements is to promote the topic not only in academia but also in industry. End users have little awareness of the potential of process mining. The vendors play a crucial role in this. During the discussion several ideas are discussed. This resulted in the following decisions and action points:

  • The Task Force will collect a set of case studies on the IEEE website. These should convince potential users that process mining is no longer experimental but can be applied in many settings. There will be a template to describe such case studies and the members will contribute.
  • There will be a “Process Mining Manifesto” by the Task Force. All task force members, but also other participants of the meeting will be invited. The goal of the manifesto is to provide a clear definition of process mining. Process mining is not limited to process discovery, but also includes conformance checking, performance diagnosis, organizational mining, prediction, recommendation, etc. The key requirement is that analysis is based on “facts” (event log) and that process models play a role (either discovered or modeled) in this.
  • The goal is to also have a blog that is open to all task for members and visible for the whole world. The blog should be broader than a particular organization/tool and visible for anyone. Christian Guenther will come with a proposal for this.
  • All members should try to effectively use the label of the Task Force in their communication. This is encouraged. At the same time Alessandro Sperduti suggests that it is wise to become a member of the IEEE Computational Intelligence Society in such situations.
  • Event logs used in academic or industrial case studies should be made publically available as much as possible, preferable using DOIs (Digital Object Identifiers), such that data can be referenced. Where applicable, anonymization should of course be applied. Boudewijn will take the lead in developing the necessary tool support and infrastructure.

The last part of the meeting was devoted to the XES format (see The format is explained and the need for a new unified format is motivated. In particular extensibility is important. All participants agree with using the format in the future. TU/e will support organizations that have problems dealing with the new standard. Note that a parser, etc. are provided. Also see the functionality of ProM 6, XESame, and Nitro. There is a clear need for managing the process of adopting extensions. The cost extension developed by Moe Wynn will be the first test case. The expectation is that the WfMC will adopt XES. The discussion on standards for event logging resulted in the following decisions and action points:

  • The task force adopts the XES standard and will use it as a tool to promote the topic.
  • Keith Swenson will discuss the topic within the WfMC. He supports the goal to replace Business Process Analytics Format (BPAF) by XES (while adding extensions if needed). There is no point in having two standards.
  • There will be a defined process for handling extensions (review cycle, request for proposals). TU/e will come with a proposal.
  • Boudewijn will write down the arguments for not using XML schema (based on discussion).

In the more open discussion the role of process mining and how to “sell it” is discussed. Joseph Carmona stresses that process mining should not be limited to business processes. There is consensus on this. At the same time, it is clear that the topic will be driven by successes and real-life applications. A final topic is the promotion of process mining by adding a competition element to the BPI workshop. Boudewijn van Dongen and Barabara Weber will come with a proposal. The goal would be a to appoint to Competition Chair and learn from other conferences where this is common. Alessandro Sperduti will provide some pointers on this.

Also see the slides for more information on the meeting. taskforceinfo-meeting-at-bpm2010.pdf

Best regards, Wil.