<p>Soil hydraulic conductivity (Ks) is a critical factor that governs the movement of water and dissolved substances through the soil, making it crucial for soil management and monitoring practices. However, due to the challenges of field measurement and its high spatial variability, this study aims to predict and map the saturated hydraulic conductivity of soils in the Beni Moussa irrigated perimeter, one of Morocco’s key agricultural areas. A total of 72 soil samples were collected at a depth of 20 cm from various agricultural plots in the region during the cropping seasons of 2020, 2021, and 2022, and were analyzed in the laboratory to determine their physicochemical properties. These properties included bulk density (BD), porosity, calcium carbonate (CaCO₃) content, organic matter (OM), organic carbon (OC), soil moisture, and soil texture, which were used as input variables for the Random Forest Regression (RFR) machine learning model. The results show that the maximum RMSE values remained below 0.25, with an average of 0.16. R² values ranged from 0.89 to 0.98, with an average of 0.94. The most significant predictors in the model were OC, OM, moisture, porosity, silt, BD, and sand, with importance analysis highlighting the influence of porosity and bulk density in Ks modeling. Accuracy and the Kappa coefficient further supported the model’s performance, with an accuracy of 99.02% and a Kappa value of 98.44%. These findings confirm the Random Forest Regression (RFR) model as a robust and reliable tool for predicting and mapping soil hydraulic conductivity in the Beni Moussa irrigated perimeter, contributing to sustainable soil management in this agricultural region.</p>