This is Flux Capacitor, the company weblog of Fluxicon.
You can find more articles here.

You should follow us on Twitter here.

BPI Challenge 2013 2

BPI Challenge 2013

Are you eager to compare your skills to that of other process miners? Do you feel like sinking your teeth into an unknown data set, test your secret weapons, and see what you can make of it?

Then the Business Process Intelligence (BPI) Challenge 2013, which has just opened, is for you! The BPI Challenge is an annual process mining competition, which takes place this year for the third time.

As Boudewijn explained in an interview last year, the trigger for the BPI Challenge was the lack of publicly available real-life data sets that could be used by process mining researchers: What do you do if you want to try out process mining, but don’t have any data yourself?

With the BPI Challenge, one real-life data set is made available for both researchers and practitioners each year. In 2011, a difficult healthcare data set was provided. In 2012, the data set was from a loan application process at a financial institution. This year, Volvo IT Belgium provided a log with events from an incident and problem management system called VINST.

The format for the BPI Challenge is an open competition, where everybody can participate. You can use any tools you want. The results need to be documented in Word or LaTeX and submitted per email by:

Deadline for submissions: July 12, 2013, 23:59 CET

Why should you participate?

Next to the honor and glory of winning the competition, of course, what is a better way to learn about process mining than to actually analyze a real-world data set? While it is possible to develop entirely new approaches to analyze the data set, the focus is on the application of process mining techniques to achieve results that are relevant from a business perspective.

Are you a researcher? This is the chance to showcase your research and test the benefit of applying it in practice.

Are you a practitioner? Last year’s winners were a data analytics team from the New York-based consultancy CKM Advisors, and they won even though they were new to the topic! The jury particularly highlighted their successful translation of analysis results into business level results and recommendations. Take a look at the submissions from last year if you would like to get an idea of how others have approached this challenge.

This year’s competition is very interesting because the data is exceptionally well-documented, and clear problems and questions have been formulated by the process owner. Head over to the BPI Challenge website now and take a look!

Analyzing the BPI Challenge data in Disco

Do you want to analyze the BPI Challenge data with Disco? Like last year, we have prepared a special Disco workspace with the challenge data, which you can fully analyze with Disco, even if you don’t own a license. Just follow these three steps:

  1. Download and install the demo version of Disco from the website here
  2. Download this BPI Challenge project file to your hard disk
  3. Open the saved project file in Disco

Screenshot Disco with BPI Challenge 2013 data

After opening the project file, you will find three data sets, for incidents, open problems, and closed problems as shown in the screenshot above.

Good luck with the challenge!

Fluxicon is a proud official sponsor of the BPI Challenge 2013. We are strong supporters of academic research and education in process mining, notably through our Academic Initiative, and our mission is to translate leading process mining research into friendly and powerful tools for professionals. We thank the organizers of the challenge for providing this bridge between academia and industry, and are delighted to be on board!

Comments (2)

[…] example, the picture below shows a screenshot of the overview statistics of the BPI Challenge 2013 data set. Can you see anything wrong with […]

[…] where you can find data sets from a hospital process (2011), a loan application process (2012), an IT Service Management process from Volvo IT (2013), and a data set from the Rabobank (2014). […]


Leave a reply