Multi-Objective Optimization in Microcontroller Selection: A Comparative Analysis of Genetic Algorithms

This paper presents a comparative analysis of multi-objective genetic algorithms for automated microcontroller selection based on user-defined parameters. Three evolutionary algorithms, namely NSGA-II, SPEA2, and MOEA/D, were adapted to the problem of selecting microcontrollers from a large database of available models, which was independently collected and verified in terms of both content and volume. Each algorithm was evaluated based on convergence speed, solution quality, stability and diversity of Pareto-optimal solutions. Experimental results demonstrate that SPEA2 achieves the best balance between accuracy and robustness, while NSGA-II offers advantages in computational efficiency. The proposed approach enables faster and more accurate component selection under dynamic market conditions.