This study aimed to model the spatial variability of soil quality in a semi-arid climate. The principal component analysis (PCA) and IDW interpolation method were used to determine and mapping the soil quality through soil properties and soil quality index (SQI) model. Topsoil samples (0-30 cm) were collected during the 2019 year at 254 points from the western area of Morocco and analyzed for 20 soil properties. The total data set (TDS) were reduced to a minimum data set (MDS) by PCA. For modeling SQI, four methods (Additive Linear A-L, Additive Non-Linear A-NL, Weighted Linear W-L, and Weighted Non-Linear W-NL) were tested and compared. Four out of 20 soil properties, including sand, EC, Fe, and Zn were retained as the independent MDS to assess soil quality. The IDW maps show irregular spatial distributions for each parameter in the area. The spatial variability maps of SQI indicate that the quality of soils in the study area is generally low (0.4 < SQI < 0.55) to moderate quality (0.55 < SQI < 0.7). The SQI values were higher for W-NL (0.498) and A-NL (0.558), while W-L (0.372) and A-L (0.378) showed similar and lower SQI values. All examined SQIs models are highly correlated with them, but only the W-SQIs (L and NL models) were correlated with sugar beet crop yield than the A-SQIs. The W-NL has the highest sensitivity index (2.731). It could be concluded that the weighted is the most sensitive and suitable method for modeling soil quality. The current study could be used as an assessment tool to help decision-making on the soil quality in Moroccan semi-arid.