In addition to capacity increase, quality also has critical importance in the cement industry. In a cement product process, the chemical properties based on the oxide composition are necessary in describing clinker characteristics. One of the most important parameters in cement product, Lime Saturation Factor (LSF) controls the ratio of alite to belite in the clinker and this factor is frequently used to evaluate the quality of cement. This study focuses on identifying LSF distribution in the site conditions. For this purpose, probabilistic (geostatistical) and non-probabilistic (neural network-based) algorithms have been used. 3D based analyses revealed some relationships in the site conditions. The accuracy studies performed by performance indicators specified that the non-probabilistic methods produced better statistical prediction capacity. Thus, the adaptive neural algorithms can ensure the results identify the quality distribution in connection with geological parameters.