Swarm intelligence optimization algorithms solve complex optimization problems by simulating the behavior of organisms in nature. This paper proposes an optimization algorithm based on the swarm behavior of mackerel, called the Mackerel Optimization Algorithm (MOA). This algorithm simulates the high-speed swimming, cooperative foraging, and rapid predator avoidance behaviors of mackerel in the ocean, introducing mechanisms such as global search, local exploitation, adaptive perturbation, and historical trajectory memory to achieve efficient multi-peak optimization search. This paper derives the algorithm's speed and position update formulas in detail, analyzes its global convergence, local exploitation capability, and swarm intelligence, and explores its application prospects in engineering optimization, high-dimensional function optimization, and machine learning parameter tuning.