Smart Sampling Strategy for Geotechnical Site Characterization

Site characterization is indispensable in geotechnical engineering practice, and it aims at delineating spatial distribution of underground soils and rocks in a project site and estimating soil properties for geotechnical analysis and design through in situ tests, laboratory tests, or other methods. In geotechnical practice, soil properties are often sparsely measured at a limited number of locations, due to time or budget limit, technical or access constraints, etc. This leads to a question of how to select the number of measurements (i.e., sample size) and their corresponding sampling/measurement locations such that as much as possible information on soil properties can be obtained. A smart sampling strategy is developed in this study that leverages on innovative data analytic methods (e.g., Bayesian compressive sensing, BCS, and information entropy) for determination of sample size and locations. Real laboratory test data are used to illustrate application of the proposed smart sampling strategy.