Optimization of Agricultural Planting Resource Allocation Based on Particle Swarm Optimization and Monte Carlo Simulation

This paper aims to optimize the allocation of agricultural planting resources and enhance agricultural production efficiency and economic benefits through data preprocessing, outlier detection, linear analysis, and a multi-variable linear programming optimization model. Initially, we preprocessed the data, including handling missing values and data visualization, to ensure data quality and the effectiveness of the models. During the data preprocessing phase, we employed the 3σ principle and boxplots for outlier detection and utilized scatter plots and the Pearson correlation coefficient to analyze the linear relationships between variables. Subsequently, we established a multi-variable linear programming model that addresses the uncertainties in expected sales volume, planting costs, yields per acre, and sales prices through Monte Carlo simulation and Particle Swarm Optimization (PSO). The objective of the model is to maximize the total revenue from crops while meeting planting and growth constraints. Through the iterative process of the Particle Swarm Optimization algorithm, we obtained the optimal planting plan and generated the corresponding convergence chart. Ultimately, this paper provides data-driven decision support for crop planting planning from 2024 to 2030.