Process Mining in production to optimize milling tool usage


 

Projektteam

J. Sasse, R. Jenke, RWU Ravensburg-Weingarten University ofApplied Science
L. Prasolb, Hilti AG

1.Problem statement

  • High costsformillingtoolsin machining (Fig. 1) and missingroot cause:
    • Notransparencyaboutreal toolwear
    • Different operatinglifetimesdue toheuristics-basedsetupandexchangeofmillingtoolsbyworkers

  • Task: Perform association analysis to determine relationships and influencing factors on tools‘ lifetime* and determin suitable measures to optimize tool usage in machining
    (*measured by parts produced during lifetime)

 

Fig.1
Fig. 1: Milling tool in action. Source: zerspanungstechnik.at

2.Data processing

Data processing

 

3.Method

Method
 

4.Results

Results


5.Future Perspective

  • Individual tool settings for each tool-part-combination to optimize tool usage
  • Use of model as data-based decision support regarding tool exchanges and tool life adjustments which could also be replenished by cost/time optimized tool changes (Fig. 3)

 

Fig.3
Fig. 3: Visualization of data-based decision support.


Literature

  1. Sasse, J. (2020): Process Mining in der Produktion –Durchführung einer Datenanalyse in der spanenden Fertigung. Master Thesis, RWU.
  2. Cleve, J. & Lämmel, U. (2016): Data Mining. De Gruyter Studium, 2nded., Berlin: De Gruyter.
  3. Kröckel, J. (2019). Data Analytics in Produktion und Logistik. 1. Auflage, Vogel

 

Projektpartner

Hilti

Hilti

 

 

Downloads

Showcase-Hilti.pdf (PDF, 811.06 KB)