Identification of geotechnical units in soil exploration through principal component analysis and clustering

• Published in: International journal for numerical and analytical methods in geomechanics 2024-04, Vol.48 (6), p.1681-1699
• Main Authors: San Roman Iturbide, Oscar, Botero Jaramillo, Eduardo
• Format: Article
• Language: English
• Subjects: Cluster analysis; Clustering; Cone penetration tests; cone penetration tests (CPT); Construction; Construction industry; Exploration; Geotechnical engineering; geotechnical unit; Logging; principal component analysis (PCA); Principal components analysis; Project engineering; Soil; Soil characteristics; Soil sampling; Soil types; standard penetration tests (SPT); Stratigraphy
• ISSN: 0363-9061; 1096-9853
• DOI: 10.1002/nag.3704

Obtaining accurate information on soil characteristics is essential for large‐scale construction projects. Geotechnical exploration methods are commonly used to obtain information on soil characteristics. However, integrating the characteristics obtained through different exploration methods can be challenging. To address this problem, this study proposes the use of Principal Component Analysis (PCA) to combine geotechnical variables obtained from cone penetration tests (CPT), standard penetration tests (SPT) with soil sampling, and Suspension PS Logging tests. PCA and clustering analysis were applied to the data collected from a study area, and the results were highly effective in identifying the stratigraphy and soil type, leading to the definition of geotechnical units. This approach provides valuable information for geotechnical engineering projects, enhancing the accuracy of stratigraphic investigations and improving our understanding of soil characteristics and their influence on construction projects. Moreover, the proposed technique can be applied in other areas and different types of soils, making it a valuable tool for evaluating soil characteristics.