BPI Challenge

The tenth International Business Process Intelligence Challenge is again co-located with ICPM. This challenge provides participants with a real-life event log, and challenges them to analyze these data using whatever techniques available, focusing on one or more of the process owner’s questions or proving other unique insights into the process(es) captured in the event log.

We strongly encourage people to use any tools, techniques, methods at their disposal. There is no need to restrict to open-source tools, and proprietary tools as well as techniques developed or implemented specifically for this challenge are welcome.

Important dates

Publication of the data: March 23, 2020
Abstract submission deadline: August 24, 2020
Report submission deadline: August 31, 2020
Presentation of the winners: At ICPM 2020 Conference
Conference dates: 5-9 October 2020, Padova, Italy

Note that the submission deadlines are strict to allow sufficient time for the Jury to read the reports and assess them. It is advised that all participants keep the conference dates free in their agenda’s to be able to accept an invitation to join the conference in case you win the challenge.


Description of the Challenge

In many organizations, staff members travel for work. They travel to customers, to conferences or to project meetings and these travels are sometimes expensive. As an employee of an organization, you do not have to pay for your own travel expenses, but the company takes care of them.

At Eindhoven University of Technology (TU/e), this is no different. The TU/e staff travels a lot to conferences or to other universities for project meetings and/or to meet up with colleagues in the field. And, as many companies, they have procedures in place for arranging the travels as well as for the reimbursement of costs.

On a high level, we distinguish two types of trips, namely domestic and international.

For domestic trips, no prior permission is needed, i.e. an employee can undertake these trips and ask for reimbursement of the costs afterwards.

For international trips, permission is needed from the supervisor. This permission is obtained by filing a travel-permit and this travel permit should be approved before making any arrangements.

To get the costs for a travel reimbursed, a claim is filed. This can be done as soon as costs are actually payed (for example for flights or conference registration fees), or within two months after the trip (for example hotel and food costs which are usually payed on the spot).


The Data

For this year’s Business Process Intelligence Challenge, we collected data from the reimbursement process at TU/e. The files contain data from 2017 (only two departments) and 2018 the full TU/e.

The data is split into travel permits and several request types, namely domestic declarations, international declarations, prepaid travel costs and requests for payment, where the latter refers to expenses which should not be related to trips (think of representation costs, hardware purchased for work, etc.).

The data is anonymized in such a way that no TU/e internal IDs are visible in the final dataset, i.e. all identifiers are freshly generated. Furthermore, the amounts mentioned in the data are not exact amounts. However, adding declarations referring to the same travel permit and then comparing them to the original budget/estimate is still possible. Furthermore, on large enough samples of the data, the summed/averaged amounts should be roughly correct.

Staff members cannot be identified in the data. Instead, for all steps, the role of the person executed the step is recorded. The resource recorded in the data is either the SYSTEM, a STAFF MEMBER or UNKNOWN, or, on occasion, the data is MISSING.

The data is available for download here: (and is available as a collection in the 4TU center for research data: https://doi.org/10.4121/uuid:52fb97d4-4588-43c9-9d04-3604d4613b51)

An explanation on the data attributes can be downloaded here: Explanation TRX records EN

The Process Flow

The various declaration documents (domestic and international declarations, pre-paid travel costs and requests for payment) all follow a similar process flow. After submission by the employee, the request is sent for approval to the travel administration. If approved, the request is then forwarded to the budget owner and after that to the supervisor. If the budget owner and supervisor are the same person, then only one of the these steps it taken. In some cases, the director also needs to approve the request.

In all cases, a rejection leads to one of two outcomes. Either the employee resubmits the request, or the employee also rejects the request.

If the approval flow has a positive result, the payment is requested and made.

The travel permits follow a slightly different flow as there is no payment involved. Instead, after all approval steps a trip can take place, indicated with an estimated start and end date. These dates are not exact travel dates, but rather estimated by the employee when the permit request is submitted. The actual travel dates are not recorded in the data, but should be close to the given dates in most cases.

After the end of a trip, an employee receives several reminders to submit a travel declaration.

After a travel permit is approved, but before the trip starts, employees can ask for a reimbursement of pre-paid travel costs. Several requests can be submitted independently of each other. After the trip ends, an international declaration can be submitted, although sometimes multiple declarations are seen for specific cases.

It’s important to realize that the process described above is the process for 2018. For 2017, there are some differences as this was a pilot year and the process changed slightly on several occasions.


Some Questions

The following questions are of interest:

  • What is the throughput of a travel declaration from submission (or closing) to paying?
  • Is there are difference in throughput between national and international trips?
  • Are there differences between clusters of declarations, for example between cost centers/departments/projects etc.?
  • What is the throughput in each of the process steps, i.e. the submission, judgement by various responsible roles and payment?
  • Where are the bottlenecks in the process of a travel declaration?
  • Where are the bottlenecks in the process of a travel permit (note that there can be mulitple requests for payment and declarations per permit)?
  • How many travel declarations get rejected in the various processing steps and how many are never approved?


Then there are more detailed questions

  • How many travel declarations are booked on projects?
  • How many corrections have been made for declarations?
  • Are there any double payments?
  • Are there declarations that were not preceded properly by an approved travel permit? Or are there even declarations for which no permit exists?
  • How many travel declarations are submitted by the traveler and how many by a mandated person?
  • How many travel declarations are first rejected because they are submitted more than 2 months after the end of a trip and are then re-submitted?
  • Is this different between departments?
  • How many travel declarations are not approved by budget holders in time (7 days) and are then automatically rerouted to supervisors?
  • Next to travel declarations, there are also requests for payments. These are specific for non-TU/e employees. Are there any TU/e employees that submitted a request for payment instead of a travel declaration?


We are aware that not all questions can be answered on this dataset and we encourage the participants to come up with new and interesting insights.


The Categories

Like last year, there are two categories, namely students and non-students.

The Student Category

This category targets Bachelor, Master and PhD students or student teams. In this category, the focus is on the originality of the results, the validity of the claims and the depth of the analysis of specific issues identified. We expect participants can focus on a specific aspect of interest and analyze this aspect in great detail. Here, one can choose for example to focus on specific models, such as control-flow models, social network models, performance models, predictive models, etc.

The Non-Student Category

This category targets academics and professionals to show their skills in analyzing business processes. The submitted reports are judged on their level of professionalism and originality of the results. The participants are expected to report on a broader range of aspects, where each aspect does not have to be developed in full detail. The report submitted in this category will be judged on its completeness of analysis and usefulness for the purpose of a real-life process mining setting.

The Prize

The winner teams in both categories will be selected by a jury and invited to present the results at the conference. They will also receive a certificate, and a small token of appreciation.


Submissions should be made through EasyChair at https://easychair.org/conferences/?conf=icpm2020 where you indicate your submission to be a BPI Challenge submission. A submission should contain a pdf report of at most 25 pages, including figures, using the LNCS/LNBIP format (http://www.springer.com/computer/lncs?SGWID=0-164-6-791344-0) specified by Springer (available for both LaTeX and MS Word). Appendices may be included, but should only support the main text.

Questions about the challenge

Like before, participants can post questions about the data/process in the ProM forum. The company monitors the messages there and will try to respond as soon as possible.


Received Submissions

We received 37 submissions by a total of 198 different authors from 11 countries. The overall winner in the professional category is

The winner in the student category is:

Runner ups in both categories are (sorted on submission time, not on quality):