2026-01-222026-01-222025-09-16SANTOS, Isaac Moraes dos. Otimização multiobjetivo de misturas de concreto utilizando xgboost e algoritmo genético: uma abordagem sustentável baseada em IA. Orientador: Edilson Morais Lima e Silva. 2025. 74 f. Trabalho de Curso (Bacharelado em Engenharia Civil) – Faculdade de Engenharia Civil, Instituto de Tecnologia, Universidade Federal do Pará, Belém, 2025. Disponível em: https://bdm.ufpa.br/handle/prefix/9114. Acesso em:.https://bdm.ufpa.br/handle/prefix/9114This work proposes an innovative approach for the multi-objective optimization of concrete mixtures, aiming to reconcile mechanical strength, cost reduction, and carbon footprint (CO₂) decrease. The methodology integrates the predictive XGBoost algorithm, with uncertainty quantification, and the NSGA-II genetic algorithm for optimization. XGBoost was refined to predict concrete strength, while NSGA-II explored solutions that balance multiple performance and sustainability objectives. The results demonstrate the effectiveness of the machine learning model in optimizing concrete mixtures, statistically and mathematically validating its precision and robustness compared to traditional methods. This study highlights the potential of artificial intelligence to enhance civil engineering, offering tools for more informed and sustainable decisions. Future steps include laboratory testing for practical validation of computational results, consolidating the applicability of the proposed approach in a real context.Acesso AbertoOtimização multiobjetivoConcretoMachine learningXGBoostAlgoritmo genéticoSustentabilidadeResistência a compressãoTraço de concretoMulti-objective optimizationConcreteMachine learningXGBoostGenetic algorithmSustainabilityCompressive strengthConcrete mix designCNPQ::ENGENHARIAS::ENGENHARIA CIVILTrabalho de Curso - Graduação - MonografiaAttribution-NonCommercial-NoDerivs 3.0 Brazil