University of Twente Student Theses

Login

Finding the Appropriate Level of Abstraction for Process Mining in Logistics

Maneschijn, D.G.J.C. (2021) Finding the Appropriate Level of Abstraction for Process Mining in Logistics.

[img] PDF
381kB
Abstract:Process mining is a field within data science that focuses on the execution of business processes, as observed from real-life events. The mined process models often contain too much detail and, therefore, become hard to understand for various stakeholders. To solve this, abstraction could be applied. Abstraction is about simplifying a process model, to make it more comprehensible. It is often hard to find the right balance between having a model that gives actionable information, but that is not too detailed. Moreover, what level of abstraction is appropriate depends on what stakeholder is looking at the process model. This research defines multiple levels of abstraction and generates process models at each of these levels. Using quantitative measurements and an expert analysis we will reason on the quality of each process model for various stakeholders. We found that there exists a level of abstraction that is deemed appropriate for all stakeholders by both the quantitative analysis and the expert analysis. This research contributes to the existing research on abstraction in process mining, by explicitly defining a set of abstraction levels for the fuzzy miner. Moreover, we present how quantitative measures in combination with an expert analysis can be used to reason on the quality of a process model considering the needs of various stakeholders. This reasoning is used to define the most suitable abstraction level for every stakeholder.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business & IT BSc (56066)
Link to this item:http://purl.utwente.nl/essays/86915
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page