MOPDO: a multi-objective prairie dog optimizer for engineering design problems
Küçük Resim Yok
Tarih
2025
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Springer Heidelberg
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
This research introduces a novel multi-objective version of the recently proposed prairie dog optimizer, the multi-objective prairie dog optimizer (MOPDO). Inspired by the foraging and burrowing activities of prairie dogs, which entail search exploration and particular responses to distinctive alarms for exploitation, MOPDO is proposed. This is a Pareto dominance-based approach to a modified and enhanced version of its single objective counterpart. MOPDO is able to deal with multiple objectives, explore, and exploit promising regions in the optimization landscape, and identify non-dominated solutions, providing decision makers with valuable trade-off choices. To demonstrate its practical applicability, MOPDO is applied to tackle five challenging structural design problems, each characterized by two conflicting objectives: two objectives, minimizing structure weight and minimizing maximum nodal displacement. The algorithm is compared against two other state-of-the-art multi-objective algorithms and rigorous evaluation is conducted using Hypervolume testing. The results show that MOPDO performs better than the comparison algorithms and is able to find a diverse set of non-dominated solutions. Statistical analysis of the experimental results using Friedman's rank test is conducted to further investigate the experimental results. MOPDO's solutions and convergence behaviour show that MOPDO is a very efficient method to solve complex design problems and is superior to the existing multi-objective algorithms in terms of effectiveness and efficiency.
Açıklama
Anahtar Kelimeler
Multi-objective, Truss design, Computational analysis, Exploitation, Exploration, Constraints techniques
Kaynak
International Journal of Interactive Designand Manufacturing - Ijidem
WoS Q Değeri
N/A
Scopus Q Değeri
Q2
