Process Mining Challenge won with LANA

Process Mining Challenge won with LANA
July 4, 2019 Shirin

Gartner has declared LANA a pioneer in Process Mining in an international comparison. LANA’s strengths lie in advanced automation through machine learning, flexible and open interfaces and efficient data transformation. Dr. Luise Pufahl and Ph.D. students Simon Remy and Kiarash Diba from Hasso Plattner Institute (HPI) tested in practice how well LANA actually delivers. And with success! With the help of LANA, the team won the BPI Challenge 2019 and demonstrated how LANA can be used to effectively perform extensive process analyses.

The BPI Challenge was first launched in 2011 as an initiative of the three Dutch technical universities. The objective was to make a real dataset publicly available so that scientists and practitioners could test current Process Mining techniques and present them to the community. Since then, the winners of the Challenge have been honored by the Business Process Community.

Process Mining is no longer just a topic for pioneering companies but has reached the masses. And for a good reason. According to Gartner, Process Mining is “one of the cornerstones of a company-wide initiative for digital transformation”. Process Mining is now being used in a wide variety of areas such as production and logistics chains, automation and IoT applications. Thanks to increasingly advanced technology, the application possibilities – and thus the demand – are increasing.

ICPM 2019 – the first international Process Mining conference

Due to the high relevance of the topic, ICPM, the international conference on Process Mining, was held for the first time this year. At ICPM 2019, experts from industry and science exchanged information on current research results, trends, and tools in the market. One of the highlights was the awarding of prizes to students, doctoral students, and specialists in the Process Mining Challenges.

The HPI team mastered the BPI Challenge with the help of LANA Process Mining. The task: to analyze the purchasing process of a multinational company from the Netherlands using Process Mining. The database provided, a so-called log file, contained more than 1.5 million events from 2018. The participants had to answer various questions, in particular regarding process conformance.

The challenge consisted of

– Creating target process models according to specific categories,
– the performance of the invoicing process
– and analyzing deviations from the target model.

The HPI team won against other Challenge participants with its excellent report supported by LANA’s analysis results. LANA has thus successfully proven itself in practical tests. Dr. Luise Pufahl was particularly enthusiastic about the possibilities of automated performance analysis: “With the Follow-Relation Filter, we were able to determine the time between any two events within seconds. For example, we were able to determine the time between the creation of the order invoice and goods receipt and thus efficiently solve part of the challenge. Another exciting feature, according to the HPI team, was the AI-based root cause analysis, with which they could automatically determine the causes of inefficiencies and deviations.

Would you like to convince yourself? Let our experts present you LANA in a demo!