Optimized Decision Making Research on Large Scale Data Based on Dynamic Risk Assessment and Particle Swarm Algorithm

This paper systematically designed an optimal crop-planting scheme by comprehensively considering market demand and mutual influence between crops with the following multi-objective function planning model and implemented it by Python programming. In this paper, the division and merging of farmland and greenhouses were preprocessed. On the basis of the analysis of the planting scheme in North China, a nonlinear planning model of profit maximization was established. By iterative solution using the particle swarm algorithm, the optimal planting scheme from 2024 to 2030 was derived. Then, the factors of risk assessment were introduced and several scenarios set up. Through five sets of experiments, a discussion on the impacts of annual growth rate, sales volume change, and cost increase was conducted on the planting scenarios. Finally, the scenario with maximum total profit was explored. Finally, through the studies on substitutability and complementarity between crops, expected sales volume, sales price and planting cost were correlated by Pearson's coefficient. Contrastive experiment was designed to get the variation rule of profit and the relationship of profit to total profit in different lands, which provided scientific basis and guiding in practice for optimizing the plant scheme.