Datengetriebe Prozessoptimierung: Einsatz von Machine Learning in flexiblen Prozessen zur Effizienzsteigerung für die Automobilproduktion

Data driven process optimization: Use of machine learning in continuous processes for automotive production

Masterthesis, Bachelorthesis, Advanced Design Project (ADP)

The innovative manufacturing process of linear flow splitting enables the resource-efficient production of branched profiles from flat sheets. Due to their geometry and manufacturing-related properties, these sheets are ideal for use in automotive and other transport applications.

The tool system used in this process is highly flexible and therefore not trivial to adjust and operate. A large number of input and output variables must be monitored and optimised. The aim of this work is to describe the linear flow splitting using a data-driven approach and to increase its efficiency. This involves drawing on a large amount of process data from an automated plant.

In addition to process optimisation, the reliable determination of product quality also plays a central role. By analysing the process data, not only can potential sources of error be identified at an early stage, but correlations between the various parameters and product quality can also be uncovered.

Research method

Experimental, Theoretisch