Selecting high-quality software components to build computer applications that simultaneously meet user needs represents a major challenge. Mathematical models and optimization tools exist in the literature to facilitate the selection of components for experimentation and to aid decision-making in complex selection processes. However, these exact methods can only identify a single optimal solution within a limited set, due to the difficulties in managing the memory of the solvers associated with these tools and the computational time required for the solutions.In this paper, we propose a heuristic approach based on genetic methods. These models and genetic algorithms allow the classification of selected components according to multiple objectives based on functional coverage, licensing, and component learning time. Our approach enables us to address component selection involving often conflicting objectives.The methodology used considers software components from various markets (ComponentSource, WordPress) and employs non-dominated sorting genetic algorithms to distribute solutions across a so-called Pareto front. Our models will help find a good compromise between different, often conflicting objectives, including decisionmaker preferences, budgetary constraints, and project deadlines.