University of Twente Student Theses
Dutch language specific KPI exploration and recommendation for maintenance scheduling
Mansvelder, F.M. (2022) Dutch language specific KPI exploration and recommendation for maintenance scheduling.
This is the latest version of this item.
PDF
4MB |
Abstract: | Little work exists on Process Mining with a focus on Key Performance Indicators. Often, KPIs are omitted entirely due to shortcomings of KPIs such as generalization. In practice, datasets might not be sufficiently structured. Many organizations maintain administrations in their native language, which gives rise to language-specific issues and challenges that render off-the-shelf Process Mining solutions unviable. In cases where a dataset is not compatible with common Process Mining practices, KPI-based strategies may prove more robust. Using one such dataset containing descriptions (un)scheduled maintenance tasks in Dutch, this thesis illustrates how KPIs can be applied where Process Mining is not viable, by dealing with the generalization problem of KPI in a Process Mining-inspired manner. By calculating KPI values within a specified context, and comparing outlying values, recommendations to improve business performance can be made without generalizing. Additionally, the challenges posed by the Dutch language and the methods used to deal with these are reported. |
Item Type: | Essay (Master) |
Clients: | De Groot, Hengelo, Nederland |
Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
Subject: | 54 computer science |
Programme: | Computer Science MSc (60300) |
Link to this item: | https://purl.utwente.nl/essays/90449 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page