Research on Crop Planting Strategy Optimization Based on Non-Monopolize Search and Catch Fish Optimization Models

Abstract

Crop planning under uncertainty is a critical issue in modern agriculture, especially when facing variable climate conditions, resource limitations, and unstable market demands. This study aims to develop an effective framework for optimizing multi-season, multi-crop planting strategies across heterogeneous land types, considering both deterministic and uncertain input conditions. In the deterministic setting, a profit-maximization model is formulated, incorporating realistic constraints such as land type, seasonal cycles, planting limitations, and crop rotation rules. The model is solved using the Catch Fish Optimization Algorithm, which demonstrates rapid convergence and stable performance. To account for uncertainties in crop yield, cost, and market price, we extend the model into a multi-scenario framework and apply the Non-Monopolize Search Algorithm to optimize expected profit and worst-case outcomes jointly. Two surplus-handling policies—complete disposal of excess crops and discounted post-harvest sales—are evaluated and compared. Experimental results show that the Non-Monopolize Search Algorithm significantly improves robustness, delivering high-quality solutions across diverse scenarios. The proposed dual-algorithm approach provides a practical and adaptable tool for agricultural decision-making and can be generalized to broader applications in resource allocation and policy evaluation under uncertainty.