A swarm-intelligent hybrid optimization framework for robust agricultural planning under uncertainty

This study presents a robust and adaptive hybrid optimization framework for agricultural crop planning under uncertainty, leveraging computational intelligence techniques to improve profitability, resilience, and decision efficiency. The proposed framework integrates linear programming, Monte Carlo simulation, and particle swarm optimization (PSO) to jointly handle economic constraints, resource limitations, and stochastic variability in yield, price, and demand. A greedy pre-allocation algorithm is employed for crops with strict field-type or seasonal constraints, simplifying global optimization without sacrificing solution quality. Monte Carlo simulation is used to assess the impact of stochastic disturbances on decision robustness, while multivariate regression and correlation analysis enrich the objective function by capturing interdependencies among crops. Comparative results demonstrate that the PSO-based model outperforms traditional deterministic methods in both profit stability and adaptability, with reduced variance and stronger robustness under fluctuating conditions. This research highlights the value of swarm intelligence in complex, data-driven agricultural decision-making, offering a scalable solution for sustainable and intelligent crop planning.