Research Projects Ongoing


Click to expand the following past research projects of the Workgroup Industry 4.0.

Research Motivation

Manufacturers face an increased global competition and products are demanded within shortening lead times, while at the same time a high delivery reliability is required
At the same time, the increasing amount of product variants leads to correlated, unstable, and highly unpredictable demands
The resulting frequent changes to the production plan and can lead to a variety of unforeseen disturbances and interdependencies between production orders
In this research, the effect of production order interdependencies on logistics performance is studied and new models for production planning are derived

Research Framework

The funnel model for logistics performance was developed as a general framework for describing production systems and deriving logistics performance indicators

Figure 1: The classic funnel model for logistics performance (Wiendahl, 1987)

This funnel model is based on an analogy of physical, granular matter systems with particles flowing through a hopper
While granular matter models descibe particle flow by taking into account particle-particle interactions, the funnel model for logistics performance does not account for interdependent effects between production orders
Hence a research framework was developed in figure 2:

Figure 2: Research framework based on Pöschl & Schwager (2004) and Wiendahl (1987)

Research Approach

To investigate order interdependencies a 3-step approach was chosen:

  1. Real data analysis: Analyse real production feedback data to show interdependency effects
  2. Simulation studies: Investigate the effects of environmental conditions on the strength of interdependencies and the resulting effect on logistics performance
  3. Develop recommendations: Derive recommendations for production planning to avoid negative effects of order interdependencies

First Results/Outlook

Analysing real data sets of production ordersusing a new analysis framework, Enrichment Analysis, shows that when deleting data for orders that pass through similar machines at roughly the same times, interdependencies can be eliminated
There is a spatiotemporal boundary to order interdependencies that needs to be investigated further
Environmental conditions which lead to interdependencies between orders need to be investigated

Figure 3: Real data analysis of order interdependencies; red curve shows a decrease in dependency effects between orders when deleting neighbouring orders

A simulation model was developed to replicate real data conditions
In a next step, different environmental conditions will be trialed and the effects between order interdependencies and key logistics performance indicators will be established

Contact

Victor Vican, M.Sc.
Research Associate
RWTH Aachen University
Lehrstuhl für das Management von Industrie 4.0
Chair of the Management of Digitalization and Automation
Kackertstraße 7
Office 142
52072 Aachen
Tel.: +49 241 80 96175

Email: vican@scm.rwth-aachen.de
Web: www.scm.rwth-aachen.de


Ph.D. Student
Department of Mathematics and Logistics
Jacobs University gGmbh
Campus Ring 1
28759 Bremen
tba
tba