Antkiewicz, MichałMyszkowski, Pawel B.Gmyrek, KonradKrzeminski, AdamCalvo-Rolle, José Luis2024-09-122024-09-09Antkiewicz, M., Myszkowski, P.B., Gmyrek, K., Krzeminski, A., Calvo-Rolle, J.L. (2024). Efficiency of Specialized Genetic Operators in Non-dominated Tournament Genetic Algorithm (NTGA2) Applied to Multi-objective Multi-skill Resource Constrained Project Scheduling Problem. In: Nguyen, NT., et al. Advances in Computational Collective Intelligence. ICCCI 2024. Communications in Computer and Information Science, vol 2166. Springer, Cham. https://doi.org/10.1007/978-3-031-70259-4_8http://hdl.handle.net/2183/39000This proceeding paper is an accepted manuscript version accepted for publication, and is subject to Springer Nature’s AM terms of use https://www.springernature.com/gp/open-research/policies/accepted-manuscript-terms. Version of Record: Antkiewicz, M., Myszkowski, P.B., Gmyrek, K., Krzeminski, A., Calvo-Rolle, J.L. (2024). Efficiency of Specialized Genetic Operators in Non-dominated Tournament Genetic Algorithm (NTGA2) Applied to Multi-objective Multi-skill Resource Constrained Project Scheduling Problem. In: Nguyen, NT., et al. Advances in Computational Collective Intelligence. ICCCI 2024. Communications in Computer and Information Science, vol 2166. Springer, Cham. https://doi.org/10.1007/978-3-031-70259-4_816th International Conference on Computational Collective Intelligence, 9-11 September 2024, Leipzig, Germany.[Abstract] The Multi-Objective Multi-Skill Resource Constrained Project Scheduling Problem (MS-RCPSP) is an NP-hard real-world problem that can be solved by metaheuristics like the Non-Dominated Tournament Genetic Algorithm (NTGA2). NTGA2 method is effective as a generic black-box metaheuristic. In the paper, we present experiments to examine how effective NTGA2 is in multi-objective optimization when the black-box rule is omitted, and specialized operators are used: Cheaper Resource Crossover, Less Assignment Crossover, and Resource-Leveling Mutation. Experimental results show that specialized operators have extra computational costs, but finally, the NTGA2 method is the most effective. Results are based on the benchmark iMOPSE library, compared to state-of-the-art methods, and statistically verified.engMulti-objective optimizationEvolutionary computationSpecialized operatorsSchedulingconference outputopen accesshttps://doi.org/10.1007/978-3-031-70259-4_8