Methods for coordinating and aggregating expert information on the uncertain parameters of aircraft components were developed. They were modified by using uncertainty theory methods, a mathematical model, and algorithms for generating aircraft configuration options. These methods use clustering into a fixed number of classes to aggregate the generated options and identify characteristic types of aircraft configurations. It reduces the dimensionality of the data during multivariate analysis of design solutions. Algorithms for constructing Pareto fronts and selecting optimal representatives for each cluster were developed based on weighted criteria aggregation. Based on the developed methods and models, a three-level architecture for transforming uncertainty into deterministic recommendations is proposed. There are a generation of non-deterministic component characteristics; a generation of configuration options; and an aggregation and clustering of options. Aircraft components are formalized as sets of uncertain parameters defined by corresponding distributions: a wing, a fuselage, an engine. The developed methods and models were validated based on a comparison of the obtained results with data about real aircraft. A formalized method and models for matching and aggregating expert information based on clustering methods for a fixed number of classes. Testing on real data, as a result of which 4 deterministic recommendations on optimal aircraft configurations were obtained. Implementation of the developed methods in the form of algorithms that can be integrated into an intelligent system, providing a choice of technologically feasible and cost-effective solutions. Clustering allows you to group aircraft configuration options, identifying characteristic types of design solutions. Verification based on real aircraft data has confirmed the adequacy and practical applicability of the method of matching and aggregating expert information in solving urgent aircraft engineering problems. At the same time, it is possible to reduce the design time by automating the processes of generation and evaluation of limited configuration options.