Abstarct: Absract In the process of designing the foundation of structures on different soils, bearing capacity and soil subsidence are considered as two main criteria by engineers, so that the construction of structures on weak soils causes destructive subsidence and instability of the structure. Turns. Therefore, the development of methods for improving poor soils is essential. Stone columns are considered as one of the suitable tools for improving poor soil. The soil sample studied in this study was sandy soil (Firoozkooh sandy soil). The main purpose of this study was large-scale numerical and laboratory modeling of the performance of the rock column group in improving the behavior of sandy soils. In order to evaluate the performance of the group of rock columns in sandy soil, reinforced sandy soil with different arrangement, number and geometric characteristics has been modeled. In this study, from three single arrangements (single stone column), triangular (three stone columns) and square (four stone columns), three numerical values per column diameter (60, 90 and 120 mm), three numerical values per Height to diameter ratio of stone columns (3, 5 and 7), three numerical values per center-to-center distance of stone columns baxsed on diameter (2, 2.5 and 3) and 2% soil density (0% and 35%) were used. Is. The responses studied in different modeling scenarios include maximum shear capacity, maximum stress and maximum displacement created in the soil bed. The results of this study show that by changing the arrangement (number of stone columns), changing the diameter of the stone columns, changing the ratio of height to diameter of the stone column, changing the distance from the center to the center of the stone columns and changing the soil density percentage, The bearing capacity of the soil, the stress created in the rock columns and the maximum displacement created in the rock column will change. For example, by increasing the number of stone columns from 1 to 3 and 5, we will see an increase of 86% and 128% of the maximum soil capacity, respectively. This trend shows a 39% and 39.8% decrease in the distributed stress response.