<p>Drying is a widely employed preservation method in food processing, aimed at extending shelf life by reducing moisture content, thereby minimizing storage and transportation costs. Despite their widespread use, cabinet-type multi-tray dryers often exhibit non-uniform airflow distribution, leading to inconsistent airflow uniformity. This study presents an optimization framework to enhance airflow uniformity by strategically repositioning supply fans, integrating Computational Fluid Dynamics (CFD) simulations with Sequential Linear Programming (SLP). An L₂₅ orthogonal array guided the CFD experiment design, and a second-order polynomial regression model was developed to correlate the airflow uniformity index (U_i) with fan positions (P₁, P₂, P₃). The optimization, constrained by fixed geometric and operational parameters including a 78.80&#xa0;mm supply gap, 45° baffle angle, 2.74&#xa0;m/s Inlet airflow speed and 315&#xa0;mm fan diameter was conducted for a 32-tray dryer (dimensions: 1420 × 650 × 1600&#xa0;mm). The regression model achieved R² = 0.9247. Incorporating trust-region techniques into the SLP framework significantly improved convergence efficiency: static and dynamic trust-region approaches reduced iterations to 15 and 8, respectively, compared to 661 without trust-region control. These methods adaptively managed step sizes, promoting faster and more stable convergence without compromising accuracy. The MATLAB-based optimization yielded optimal fan positions of P₁ = 359.05&#xa0;mm, P₂ = 395&#xa0;mm, and P₃ = 395&#xa0;mm, achieving a maximum uniformity of 76.5% with 0.2% prediction error. This CFD–SLP integrated approach offers a reliable, scalable, and cost-effective solution for enhancing airflow uniformity in industrial dryers and holds promise for broader applications such as thermal or fluid process optimization tasks across food, pharmaceutical, and chemical industries.</p>