Multi-Period Crop Planting Strategy Optimization Based on Multi-Objective Programming and Monte Carlo Simulation

Abstract

This study addresses the optimization of crop planting arrangements for the period from 2024 to 2030, with the objective of maximizing the cumulative economic return. A dynamic programming framework is adopted as the primary methodology, while the decision-making logic of the greedy algorithm is incorporated to update the planting strategy year by year. Specifically, the planting decision for each year is determined on the basis of the forecast results and operational outcomes of the preceding year. For the two possible disposal strategies of unsold agricultural products, separate multi-stage optimization models are established and solved, thereby generating corresponding optimal planting plans for the entire planning horizon.In addition, the model takes into account a range of uncertainties associated with agricultural production and market conditions, including historical sales performance, expected demand for different crops, yield per unit area, cultivation costs, and fluctuations in selling prices. These factors are assumed to vary over time rather than remain constant. To evaluate the influence of such uncertainties, Monte Carlo simulation and probabilistic statistical analysis are employed to generate numerous possible scenarios. The expected performance of each planting strategy is then estimated through repeated random sampling, which improves the robustness and risk resistance of the final decision scheme.