RESUMO
The dynamic replacement method is used to strengthen the subgrade of objects, usually up to 5 to 6 m thick. After the improvement process, acceptance tests in the form of load testing are carried out. Interpretation of the test results can cause some difficulties. Dynamic replacement results in a situation where columns of different shapes, loaded with plates of diameters usually smaller than the head diameter and in the vicinity of adjacent columns, are subjected to load tests. In order to demonstrate the influence of these factors, a spatial model of soil strengthened by dynamic replacement, comprising four material zones, was calibrated on the basis of load testing. The following models were used in the analysis: linear-elastic, elastic-perfectly plastic (Coulomb-Mohr) and elastic-plastic with isotropic hardening (Modified Cam-Clay). This formed the basis for 105 numerical models, which took into account the actual shapes of the columns made at various spacings, subjected to load tests with plates of various diameters. The analyses of the settlements, calculated moduli and stress distribution in the loaded system showed how the results were significantly influenced by mentioned factors. This implies that the interpretation of the results of load tests should be based on advanced spatial numerical analyses, using appropriate constitutive models and including the considered factors.