Prediction of masonry strength and earthquake resistance of historic buildings in Vienna using machine-learning algorithm

Abstract

“Gründerzeit houses“ in Vienna/Austria, which were built between 1848 and 1918, are an essential part of the Viennese building infrastructure with a number of approx. 30,000. Since 1992, the masonry properties of over 200 buildings have been determined and evaluated by the authors. These investigations were carried out to calculate the load-bearing capacity and the seismic strength for the extension and renovation. A comprehen-sive data set was created containing the coordinates of the buildings, district, construction period, use, build-ing type, number of storeys, type of test, moisture, compressive strength of the brick, mortar and masonry, etc. This data set consists of around 40,000 independent numerical and categorical values. It was analyzed for clusters and correlations using machine learning algorithms. Among others, the unsupervised learning algo-rithms “Hierarchical Clustering”, “k-means” and the dimension-reducing algorithm “t-SNE” are used. The cor-relations found are statistically evaluated and visualized. The most important results are presented.