Szkoła Eksploatacji Podziemnej

As science progresses, the ability to predict outcomes of complex processes through mathematical functions becomes increasingly crucial. The complexity inherent in geosciences has necessitated a shift towards data-driven analysis, with methods such as kernel techniques and random forests becoming prevalent in addressing remote sensing challenges. In the realm of subsidence engineering, an empirical approach that demands a minimal number of parameters has been favored for its high precision. This approach is under continuous enhancement through the application of data science, notably statistical inference and machine learning. Currently the solution of complex mathematical tasks, like parameter estimation, is solved through the referencing of specific Python libraries. Nonetheless, a deep understanding of these data science methodologies and the functions they employ, alongside their applicability to real-world scenarios, is indispensable for accurately interpreting results. Considering the example of the Salt Cavern in the North of Germany there will be a demonstration of data-science application in the field of subsidence engineering with a focus on the associated problems.
It tackles the intricacies of parameter estimation within subsidence engineering, offering a comprehensive examination of prediction methods and variables in predicting subsidence.