The aim of the current study was to assess the spatial variability of soil quality in a semi-arid area using an indexing approach based on soil parameters. A total of 1511 soil samples (0-30 cm) were collected in 2019 and analyzed for 20 soil properties. Four soil quality indexes (SQIs) were computed using Linear (L) and Non-Linear (NL) models through Additive (A) and Weighted (W) approaches. Principal component analysis (PCA) identified four principal components (PCs) with eigenvalues > 1, accounting for a total of 64.47% of the variance in the data. The total data set (TDS) was reduced by PCA, resulting in four soil parameters (Sand, Mn, EC, and Zn) as the independent minimum data set (MDS). IDW interpolation revealed an irregular distribution for each parameter individually, but the maps created based on SQI demonstrate that soil quality is generally low (SQI < 0.55) to moderate (0.55 < SQI < 0.7). The NL model shows the highest SQI values and sensitivity index, and the statistical results show that all the SQI models tested have strong relationships with crop yields and each other. However, W-NL shows the highest correlation (r = 0.306, p < 0.01). Thus, the weighted approach is the suitable method for evaluating soil quality in Moroccan semi-arid regions and aiding decision-making.