The article deals with the description of the methodology for calculating the reproduction value of buildings. It is a mathematical model that uses data obtained from a statistical survey and calculates the total reproduction value of buildings from a large data set. The model works with many variable inputs that describe the valued building. These include, for example, the use of the building, the enclosed space of the building, the floor area of the building, the building material of the supporting structures of the building, the age of the building, the technical condition of the building, the year the building was acquired and others. Based on the results, the model was calibrated and its parameters optimized to achieve the most accurate values.