In the United States, Farm Bill programs require ranking soils according to their potential productivity. The soil survey database provides data for a wide array of soil, climate, and landscape conditions. These data were used to create a model that arrays soils by their inherent potential productivity for commodity crop growth. A consistent, natural, defensible, and timely method of arraying soils according to their potential productivity is needed for land evaluation and risk assessment. The interpretations module of the soil survey database system uses fuzzy logic to allow soils to be considered in terms of their degree of membership in the set of soils that are suitable for a particular land use. A statement can be made such as “A soil that has a given set of characteristics is a non-member, partial member, or a full member of the set of soils having high inherent productivity”. The actual linkage between a soil characteristic and the degree of membership in the set of productive soils is based on a graphed function that describes the fuzzy set. The shape of the relationship can be specified to reflect the effect of an independent variable on a dependent variable, whether it is linear, sigmoidal, bell-shaped, or any other shape based on empirical evidence. A small subset of soil survey areas located in the geographic extent of non-irrigated corn (Zea mays) growth in the US was selected to establish the basic relationships of soil, landscape, and climate properties to soil productivity. Soil properties are divided into physical and chemical characteristics. Physical properties include available water holding capacity, organic matter content, saturated hydraulic conductivity, bulk density, linear extensibility, rock fragment content, and rooting zone depth. Chemical properties include cation exchange capacity, pH, sodium absorption ratio, gypsum content, electrical conductivity, and calcium carbonate equivalency. Landscape factors include the slope gradient, depth to saturation, erosion, surface fragments, flooding, and ponding. Climatic properties include frost-free days, mean annual air temperature, and mean annual precipitation. The database was queried for soil, landscape, and climate characteristics and their values related to the yields of commodity crops. In this example, field corn yield is used as the indicator of soil productivity. Spline functions were used to explore shape of the relationships between these variables and the yield to assign minimum, optimum, and maximum levels of each variable, as required. These curves are then inserted into the fuzzy logic module of the database system. Characteristics statistically shown to be more highly correlated with productivity are given more weight in the model. Available water holding capacity, mean annual precipitation, saturated hydraulic conductivity, cation exchange capacity, and pH were shown to strongly influence productivity. The model itself reflects the interaction of the variables influencing productivity. A low score on any parameter forces a lower productivity index.