The work is devoted to solving the problem of combining heterogeneous relational databases based on integration models of different subject domains. The paper proposes methods for analyzing objects and their properties when combining models of subject domains, a method of combining integration models of different subject domains based on consistent rank evaluations of objects and the values of their typed essential properties. The model of the subject domain object is improved, which, unlike the classical one, takes into account the integration components that are important for combining: the sets of values of consistent ranks of properties and the sets of typed essential and non-essential properties of the object and their values determined on the basis of them. The subject domain model has been improved, which, unlike the existing one, takes into account certain combining scenarios and consistent ranking assessments of objects. Based on the proposed models and methods, an information technology for combining relational heterogeneous databases has been developed, which has increased the reliability of detection of subject domain objects and their properties to be combined, while simultaneously reducing the number of comparison operations for automated creation of a combined integration model of the subject domain.