Animal feed is an essential element of modern livestock production, and optimizing feed formulations is vital for balancing nutritional value, cost efficiency, and environmental sustainability. Traditional methods primarily rely on empirical knowledge or basic linear programming, often falling short of achieving this objective. To address this, this study proposes a novel hybrid method that combines genetic algorithm with hill climbing (GAHC). The approach incorporates multi-population parallel evolution, hill climbing, and simulated annealing, while also enhancing selection, adaptive crossover, and mutation operations. This integration aims to optimize animal feed formulations by improving nutritional value and reducing costs. MATLAB simulation results demonstrate that GAHC exhibits exceptional performance across three benchmark test functions and feed formulation optimization. The proposed GAHC achieves an average convergence rate of 96.67% and cost reductions of 3.7%, 2.2%, 2.8%, 1.3%, and 0.7% compared to standard genetic algorithm (GA), adaptive genetic algorithm (AGA), hill climbing genetic algorithm (HCGA), multi-population genetic algorithm (MPGA), and multi-population hill climbing genetic algorithm (MPHC-GA), respectively, significantly outperforming GA, AGA, HCGA, MPGA, and MPHC-GA.