The increasing complexity of the decision-making process, the complexity of the relationships between factors influencing the choice of one or another solution, the increase in the cost of an incorrect decision, and the lack of deterministic information for decision-making determine the relevance of the topic of the work. Modern economic systems, which change rapidly over time and have new features, require the inclusion of qualitative components and expert judgments in the decision-making model. The purpose of the work is the further development of a methodology that using a systems approach will increase the reliability of decision support problem solutions in weakly structured and unstructured complex systems based on hierarchical models. The methodology include evaluation and increasing of the consistency level of interval multiplicative pairwise comparison matrices of model elements; prioritization methods and criteria for evaluating accuracy of interval weights; aggregation of local weights of decision alternatives in terms of decision criteria when decision criteria are mutually interdependent.