Uncertainty and noise are frequently-encountered obstacles in real-world applications of numerical optimization. The practice of optimization that deals with uncertainties and noise is commonly... Show moreUncertainty and noise are frequently-encountered obstacles in real-world applications of numerical optimization. The practice of optimization that deals with uncertainties and noise is commonly referred to as robust optimization. This thesis concentrates on robust optimization w.r.t the parametric uncertaintiesin the search variables. These parametric uncertainties are assumed to be structurally symmetric, additive in nature, and can be modeled in a deterministic or aprobabilistic fashion. This dissertation empirically studies the models, algorithms, and techniques utilized for surrogate-assisted robust optimization in this context. Based on the studies performed in the dissertation, we conclude that Kriging, SVM, and Polynomial Regression are useful modeling techniques to solve robust optimization problems. We also validate the applicability of Autoencoders and PCA for addressing high-dimensional problems. Lastly, we find that mini-max robustness is the most efficient robustness formulation technique in practical scenarios. Show less
Industrial manufacturing processes, such as the production of steel or the stamping of car body parts, are complex semi-batch processes with many process steps, machine parameters and quality... Show moreIndustrial manufacturing processes, such as the production of steel or the stamping of car body parts, are complex semi-batch processes with many process steps, machine parameters and quality indicators. To optimize these complex processes, for example by reducing the number of defects or increasing the throughput, a great number of requirements need to be taken into consideration. In this dissertation a framework for monitoring and optimizing these complex industrial processes is presented. The framework is specifically tailored to the production processes of Tata Steel and BMW Group. Both are industrial partners of the PROMIMOOC project. The framework consists of several components of which; preprocessing, outlier detection, predictive modeling and optimization are the main technical components that are the focus of this work. For each of these components a possible implementation is proposed and the challenges in implementing these components in an industrial manufacturing setting are discussed Show less
