Development of a Computer Program for Geotechnical Assessment

Abstract

Geotechnical investigation is an important site characterization process for identifying suitability of soil and rock as construction materials or foundation support for engineering structures. 2D representation of geotechnical investigations such as borehole logs or geophysical sounding maps are commonly used although they are widely known to have numerous limitations that lowers quality of the investigations and subsequent risk to construction projects. There is increasing demand for 3D representation as an alternative for improved visualization of geotechnical investigations. The aim of this study was development of a computer program (geomod) for 3D representation of geotechnical investigations by integrating pit/borehole logs, topographic and geologic information, and spatial modelling. The R code for 3D modelling was developed using R language in R software (R 3.4.1) downloaded from https://cran.r-project.org/bin/windows/base/ in November 2016. The R code was done sequentially starting with the loading of libraries suitable for the modelling and data analysis and visualization followed by the import of the geotechnical data (SPT and Atterberg limits) and the predictors (soil maps). The R codes were then integrated into R package known as geomod. It’s an add-on package of R software and is available at R software website under downloadable packages (https://cran.r-project.org/package=geomod, accessed on 20 August 2023). This study used data collected from drilled boreholes(14No.) where in situ observations (e.g. SPT) were collected and also collected soil samples for lab analysis (Atterberg). The geotechnical data for the study was obtained in Kibarani area, Mombasa City, Kenya. The remote sensing images, geology map, and elevation map downloaded from USGS are some of the secondary data used in the study. The dataset is available in Geomod computer program and contains 152 points of soil horizons for the geotechnical investigation. Geomod R package was applied using the geotechnical data obtained from the study area to produce 3D model. The 3D model characterized both horizontal and vertical variations of the soil properties in one graphical illustration.