 Process Mining in Textile Production: Insights from Penn Textile SolutionsTobias Brockhoff (),
Merih Seran Uysal (),
Anahita Farhang Ghahfarokhi,
Leon Reinsch,
Thomas Kordtokrax (),
Andreas Meister (),
Franz Schütte (),
Tugsan Vural (),
Mahsa Pourbafrani (),
Thomas Gries () and
Wil M. P. van der Aalst ()
A chapter in Business Process Management Cases Vol. 3, 2025, pp 87-103 from Springer Abstract: Abstract (a) Situation faced: In the textile industry, each individual manufacturing step is typically highly optimized. Nevertheless, inter-manufacturing step dependencies usually have great potential for further optimization. Penn Textile Solutions GmbH collects its manufacturing event data in a dedicated database on the completion of each manufacturing step. However, the data are not used to generate a holistic view of the process. In this study, we apply process mining to leverage the data, generate insights, and visualize a manufacturing production process that is focused on a single machine—the tenter frame. (b) Action taken: We first focused on measuring working hour-aware time intervals (i.e., not considering off days and holidays). Then we split the analysis phase into two major parts—the manufacturing process and the quality control process. We then analyzed both parts using process models as a structuring element. For the manufacturing process analysis, we first conducted a working hour-aware analysis of lead times, process times, and machine utilization. Using a hybrid discovery approach, we discovered a process model at the machine level, used it to define production stages, and investigated the latter in more detail. Finally, we analyzed the quality process based on a model-induced case classification. (c) Results achieved: We successfully created a process model that described the manufacturing process well. Using this model, we compared cases within and between production stages. While we were able to identify critical stages, the analysis revealed significant variance that was not straightforward to explain, even when taking the impact of COVID-19 into account. The accompanying analyses of resource load, idle times, process time, and lead times were an initial means of making machine utilization accessible. Our results sparked discussions, and by comparing results, we were able to identify the bottlenecks. (d) Lessons learned: Closely integrating stakeholders helped us define realistic, realizable goals. While presenting intermediate results increased trust and understanding in the subsequent stages, it also led to interesting, unexpected discussions. Regarding the application of process mining, we found it helpful to structure the analysis using process models. However, we also recognized the limitations of existing techniques to analyze flexible processes at a detailed level. A major challenge we overcame was how to map the process in a way that took the production context into account so that we could better classify the results. Date: 2025 There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it. Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-031-80793-0_7 Ordering information: This item can be ordered from DOI: 10.1007/978-3-031-80793-0_7 Access Statistics for this chapter More chapters in Springer Books from Springer |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.