In this master's thesis, we analyzed the impact of stochastic phenomena in key stages of low-volume automotive production on the overall efficiency and robustness of the system. A digital twin approach was used, applying precise statistical distributions (normal, log-normal, triangular, exponential) to model processes such as injection molding, visual inspection, packaging, and transport. Simulation results revealed that unpredictable disruptions in manual phases pose the greatest risk for bottlenecks and work-in-progress accumulation, while automated processes ensure greater stability and throughput. Optimizing and automating the most variable phases proved to be essential for improving production efficiency, reliability, and competitiveness.
|
