Author(s)

Yongqi Chen, Sijie Ma

Corresponding Author

Yongqi Chen

Abstract

Agricultural planting optimization under limited land resources and market uncertainties represents a critical challenge in rural revitalization. This paper proposes a novel optimization approach combining parallel stochastic programming with an improved Slime Mould Algorithm (SMA). The methodology introduces a two-subsystem parallel structure to reduce computational complexity, and incorporates the newsvendor model concept to handle price uncertainties. The SMA is enhanced through Good Lattice Points initialization and lens imaging opposition-based learning strategies. Experimental results demonstrate that the improved algorithm achieves superior convergence and solution quality compared to traditional methods. The optimization model generates significant economic benefits, with the discounted sales scenario showing a 58% profit increase. The parallel optimization framework effectively balances computational efficiency and solution quality, while the improved SMA shows enhanced exploration and exploitation capabilities. This research provides practical guidance for agricultural planning and demonstrates the effectiveness of the proposed method in handling complex crop planting optimization problems.

Keywords

Crop Planting Optimization, Parallel Stochastic Programming, Slime Mould Algorithm, Good Lattice Points, Opposition-Based Learning, Agricultural Planning