13th International Workshop on Business Process Intelligence 2017

to be held in conjunction with BPM 2017 Barcelona, Spain, September 10 - 15, 2017


|

Business Process Intelligence Challenge (BPIC)

Seventh International Business Process Intelligence Challenge (BPIC’17)

In this challenge, sponsored by Minit and Celonis, we provide participants with a real-life event log, and we ask 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 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.

Our sponsors provide access to their tools for use with the BPI Challenge dataset. If you would like to use Celonis on this data, please contact them directly on BPI2017@celonis.com. If you would like to try minit on this dataset, please contact minit on BPI2017@minitlabs.com.

The Winners

In total, the challenge received 3 academic, 6 professional and 14 student submissions. The Jury was impressed by the high quality of the submissions overall. All reports will appear on this website in a few weeks after the authors get a chance to submit a final version based on the jury comments. This year's winners were all contacted in time and the sponsors made it possible for all of them to join the BPI workshop in Barcelona where we had a small award ceremony and the three winners presented their main findings.

  • Student Category: Elizaveta Povalyaeva, Ismail Khamitov and Artyom Fomenko from the Moscow Higher School of Economics, with their submission entitled: Density Analysis of the Interaction With Client (report)
  • Professional Category: Liese Blevi, Julie Robbrecht and Lucie Delporte from KPMG Technology Advisory, Belgium, with their submission entitled: Process mining on the loan application process of a Dutch Financial Institute (report)
  • Academic Category: Ariane Rodrigues, Cassio Almeida, Daniel Saraiva, Felipe Moreira, Georges Spyrides, Guilherme Varela, Gustavo Krieger, Igor Peres, Leila Dantas, Mauricio Lana, Odair Alves, Rafael França, Ricardo Neira, Sonia Gonzalez, William Fernandes, Simone Barbosa, Marcus Poggi, and Hélio Lopes from Pontifícia Universidade, with their submission entitled: STAIRWAY TO VALUE: Mining the loan application process (report)

We would like to thank the Jury for their outstanding work reviewing all submissions. About the winning submissions, the Jury remarked:

  • The authors not only observe, but also interpret the results and offer concise recommendations and suggestions for process improvement. This is done based on the evidence derived from the data, while at the same time, making clear where specific conclusions cannot be drawn due to a lack of further insights/data. This is very convincing.
  • In order to obtain a process model the authors manually created and improved a process model, switching back-and-forth between editing and analysing the replay results. This resulted in a very good and detailed model that nicely aligns with the data. 
  • The report demonstrates how mature the process mining tool landscape has become and squeezes out a lot of information through these tools.

We hope to see you all at next year's BPI workshop which will be organized in Sydney, Australia. We will again organize a BPI Challenge and of course we are looking forward to the same volume of high quality submissions.

Important Dates

Abstract submission deadline: 2 June 2017
Report submission deadline: 9 June 2017
Announcement of winners: at the 13th Workshop on Business Process Intelligence (BPI 17),
Barcelona, Spain

The Challenge

In 2012, the BPI challenge used data from a financial institute. Data that has since been used by many researchers for various papers. Due to the success of this dataset, we felt an update was in order. A new workflow system has been implemented in the company, advice from the BPI challenge 2012 has been implemented and (due to the financial crisis) the case volume has gone up considerably.

The company is particularly interested in answers to the following questions:

  1. What are the throughput times per part of the process, in particular the difference between the time spent in the company's systems waiting for processing by a user and the time spent waiting on input from the applicant as this is currently unclear,
  2. What is the influence on the frequency of incompleteness to the final outcome. The hypothesis here is that if applicants are confronted with more requests for completion, they are more likely to not accept the final offer,
  3. How many customers ask for more than one offer (where it matters if these offers are asked for in a single conversation or in multiple conversations)? How does the conversion compare between applicants for whom a single offer is made and applicants for whom multiple offers are made?
  4. Any other interesting trends, dependencies etc.

This year, for the first time, we have decided to have three categories, namely students, academics and professionals. Thanks to the sponsoring of both Minit and Celonis, we can invite winners in all three categories to join the workshop in Barcelona to present their findings.

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 Academic Category

This category targets academics. The focus in this category is much more on the novelty of the techniques applied than the actual results. This provides a great opportunity for BPI researchers to show the practical applicability of their tools and/or techniques on real-life data.

The Professional Category

This category targets professionals to show their skills in analyzing business processes. The submitted reports are judged on their level of professionalism. 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 business improvement setting.

A jury will decide which report is best and the winning participant in each category will be invited to come to Barcelona to receive a prize and to present their findings!

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. Both sponsors make their tools available for use with the BPI challenge data. Information on how to contact them will follow soon.

The Data

We are proud to say that the data for this year's BPI challenge is provided by the same company as BPI 2012. The same process is considered, five years later!

The dataset provided by the company this time is richer than before. An important difference is that the company switched systems and now supports multiple offers for a single application (in contrast to 2012, where a work-around was clearly visible in the log).

The event log provided this year contains all applications filed in 2016, and their subsequent handling up to February 2nd 2017. In total, there are 1,202,267 events pertaining to 31,509 loan applications. For these applications, a total of 42,995 offers were created. As in 2012, we have three types of events, namely Application state changes, Offer state changes and Workflow events. There are 149 originators in the data, i.e. employees or systems of the company.

For all applications, the following data is available:

  • Requested load amount (in Euro),
  • The application type,
  • The reason the loan was applied for (LoanGoal), and
  • An application ID.

For all offers, the following data is available:

  • An offer ID,
  • The offered amount,
  • The initial withdrawal amount,
  • The number of payback terms agreed to,
  • The monthly costs,
  • The creditscore of the customer,
  • The employee who created the offer,
  • Whether the offer was selected, and
  • Whether the offer was accepted by the customer.

Next to this information, many events are recorded in the log. For each (uniquely identifiable) event, the employee who caused the event is recorded, as well as a timestamp and lifecycle information. The latter is provided both in the form of the standard XES lifecycle as well as the internally used lifecycle events.

Of course, the data is fully anonymized. However, the company can map the IDs from the public event log to their own system IDs, down to the event level.

The data is provided in two files:

  1. The Application event log. This event log contains all events with the application as the case ID. Any event related to an offer also refers to an OfferID.
  2. The Offer event log. This event log contains all events related to offers, with these offers as case ID. For each offer, a corresponding application is available.

Please note that there may be multiple offers per application. However, at most one of them should always be accepted.

The XES log files are strictly conforming to the IEEE XES standard and can be loaded in any tool that is compliant with XES.

Questions about the data

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

Submissions

Submissions should be made through EasyChair at https://www.easychair.org/conferences/?conf=bpi2017 where you indicate your submission to be a challenge submission. A submission should contain a pdf report of at most 30 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.