**1. Introduction**

The present chapter describes numerical simulation of sand geomaterial with the discrete element method (DEM), based on input data acquired from physical experiments. First of all, sandy soil is composed of different shape and size particles. That means, it is necessary to provide sieving test to determine granulometry curve of investigated sand. After this step, it is used scanning electronic microscope (SEM) to determine main sand particles morphology parame‐ ters, namely area, perimeter, form coefficient, angularity, roundness, circularity, and spherici‐ ty coefficients, respectively. Next physical experiments level—sand compression with oedometer. Having all this information, it is possible to start numerical DEM modelling of sand compression test.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Granulometric curve and separate particle shape discretization are one of the most relevant existing problems for numerical modelling via DEM. There are no direct recommendations for a single‐particle subscription with spheres (single‐particle subscription level based on spheres size and quantity). Subscribing soil particles with DEM, it is very important to choose size, shape, and physical properties of modelled particles. It is possible to model realistic size oedometer device without not scaled particles if quantity of modelled particles is small. Modelling of soil which is subscribed with a big quantity of particles is not possible because of the limitations of computer calculation capacity. All simplifications are accepted in order to decrees the calculation time. Another relevant problem, arising from all accepted simplifica‐ tions, is a very small quantity of the authors working on experimental and numerical testing validation. This fact only proves that there are still a lot of problems related to dispersive systems, which are modelled with DEM.
