Methodology for Value Stream Design in Production Networks - Simulation of Material Flow and Operating Concept in the Production of Electrical Drive Components

Initial situation:

With the transition to electromobility, automotive manufacturers are faced with the challenge of rebuilding the production of electric drives and their components. At the same time, the transition to electromobility is subject to constant uncertainty regarding the technologies used and the expected demand for parts to be produced. New production systems that are set up to produce electric drives and their components must also be integrated into the existing production network.

This results in a complex system in which fundamental questions from value stream design are difficult to evaluate. What are the ideal decoupling strategies between the individual local process steps or those distributed in the production network? How can the interaction of production systems across several locations be optimally designed in terms of value stream? How does this interaction behave under increasing and fluctuating demand?

However, a comprehensive evaluation and consideration of these questions are crucial for efficient production within the production network. Suboptimal decisions in value stream design in the production network can lead to significant problems as demand and the number of variants increases in the future.

Objective:

The research project aims to develop a methodology for designing efficient value streams for producing electric drive components in production networks, considering the operational concept. With the help of the methodology, recommendations for the improvement of currently existing and future planned value streams in production networks are to be derived.

Approach:

The basis of the methodology is the use of simulation models that can map the value stream in the production network and its operational concept on an operational level. These simulation models are then used to generate result datasets for many alternative value streams and thus build up a broad database. This can be used used for further data analyses. Different data analysis methods (process and classical data mining approaches) are combined to comprehensively understand the alternative total value streams.

As a result, value stream design issues in production networks can be evaluated in existing and future planned production systems, and recommendations for improving the value stream can be generated.

Acknowledgments:

The project is sponsored by Bayrische Motoren Werke AG (BMW AG). We would like to thank BMW AG for their excellent support, cooperation, and project funding.