Msc Thesis - Des-Driven Bottleneck Analysis Of An Axle Paint Shop Using Moo & Machine Learning
Södertälje, Stockholm County, Sweden
Master thesis conducted with Uppsala University and Scania.Target and assignmentsThe primary objective of this research is to identify bottlenecks and optimize parameters in the process that affect throughput and energy consumption by using machine learning-infused simulation tools.This includes:• Process understanding and data collection.• Model development: Design a simulation model and validate it with production data.• Metamodeling: development, experimentation and performance evaluation of various machine learning algorithms for simulation metamodeling.• Parameter tuning: experimentation on how the importance scores extracted from the meta-modelling can be used for the parameter tuning of the optimization algorithm for bottleneck analysis.• Insight generation and demo: Conduct bottleneck analysis and analyze the model results to gain valuable insights into potential areas for optimization. A visualization of the results to aid the engineers in the manufacturing area is desirable.