This study analyzed spatial soil fertility variability in a 1440 m² plot in Mosquera, Colombia, to create homogeneous management zones for precision agriculture.  480 soil samples were collected using a 3x1 m grid, analyzing pH, electrical conductivity, phosphorus, exchangeable cations, microelements, and soil organic matter (SOM).  Principal Component Analysis (PCA) identified SOM, pH, and electrical conductivity as key indicators for zoning. Kriging interpolation mapped these properties, revealing high variability.  The exponential model best represented the semivariograms.  Fuzzy clustering, based on indicator thresholds, divided the plot into two zones, with high overlap between pH and SOM-based divisions.  A QUEFTS model simulated crop yield, showing that optimized N and K fertilization, based on zoning, maximized yields. The study demonstrates the effectiv...