Dodo Bird Optimization Algorithm

Swarm intelligence optimization algorithms, by simulating group behavior in nature to complete complex optimization tasks, have been widely applied in continuous optimization, engineering parameter optimization, and machine learning model tuning. However, most existing algorithms heavily rely on random perturbations or leader mechanisms during the search process, easily leading to problems such as search trajectory oscillations, rapid population convergence, and loss of diversity. To address these shortcomings, this paper proposes a novel swarm intelligence optimization method—the Dodo optimization algorithm. Inspired by the foraging behavior and survival characteristics of the dodo, this algorithm introduces a memory degradation mechanism, an island carrying capacity regulation mechanism, and a directional transition strategy driven by extinction pressure, reconstructing the search process from three levels: individual memory, population distribution, and overall evolutionary state. By constructing a degenerate individual memory model, carrying capacity regulation based on local density, and an orthogonal directional transition mechanism, the algorithm significantly enhances its ability to escape local optima while maintaining search stability. This paper systematically presents the algorithm's mathematical modeling process, iterative mechanism, and theoretical characteristic analysis, providing a complete methodological foundation for subsequent applications and extensions.