The purpose of the work is to create a complex of basic models for predicting the primary physicochemical and thermophysical properties of metal melts for the targeted formation of a high-quality metal product and increasing its competitiveness. The basis for modeling is the original concept of directional chemical bonding, the core of which is the consideration of metal melts as chemically unified systems, rather than a mechanical mixture of constituent elements, and taking into account the contribution of all components, even in small concentrations. The work uses an important information component, which is a database on the properties of metallurgical melts, which is continuously updated with modern data and contains the results of own and industrial experimental research and literature search (articles, patents, inventions, scientific developments, monographs). The importance of databases is indisputable and requires them to be brought to the inter-branch and inter-university level with open access, as a separate instance for promoting the development of the scientific level and the capabilities of scientists. Adequate mathematical models were developed based on integral parameters of interatomic interaction, and additional consideration of micro-heterogeneity parameters in their structural formation ensured high accuracy of operational forecast (R2≥0.9). A comparative analysis of the obtained melting results with the corresponding calculations based on the JMatPro software complex confirmed the effectiveness of using the interatomic interaction parameters as model ones. The results of the research are recommended for use in industrial conditions for the purpose of targeted formation of the composition and properties of smelting products, as well as reducing energy costs and reducing defects due to the adoption of operational management technological decisions with the help of the integration of the developed models in the automated control systems of technological processes of steelmaking.