Building and road constructions require previous in-situ measurements of soil properties to ensure safe building structures. However, post-construction deformations on roads can still occur and are thus monitored by Differential InSAR techniques. As a new approach, this study tries to establish a direct link between the in-situ and deformation measurements by developing a fully data-driven methodology which models road deformations based on loading/unloading conditions and soil properties. For this, relevant features are extracted from openly available datasets and analysed with machine learning algorithms to model this relationship. As a result, the Pearson correlation and coefficient of determination between soil properties, loading/unloading and the linear rate of deformation are 0.6 and 0.4 respectively. Concludingly, the resulting models with different algorithms and different sets of features are moderately accurate. The uncertainty of the models is due to three main reasons: 1. The complexity of the study area in terms of construction history 2. Lack of other necessary data 3. The uncertainties caused by the proposed methodology.
- Master thesis 'Spatial and temporal analysis of road deformation based on remote sensing and subsurface exploration'