A Study on Multi-Objective Decision Optimization Based on the NSGA-II Algorithm

This paper addresses the optimization of planting decisions using advanced multi-objective optimization models. It proposes a strategic planting plan for the period 2024-2030 with the goal of maximizing profits while accounting for fluctuations in market prices and crop yields. Monte Carlo simulations are employed to model the inherent uncertainties, generating a range of possible scenarios under different conditions. A multi-matrix genetic algorithm is used to search for optimal planting strategies, allowing for efficient exploration of large solution spaces. The results indicate annual profits of 3 million and 5.6 million under various scenarios, demonstrating the potential for improved productivity and profitability when uncertainty is effectively incorporated into the model. To address the volatility in prices and yields, a robust optimization model is introduced. This model incorporates variations in crop prices and yields, improving decision-making under uncertainty. By integrating the NSGA-II algorithm, the model is able to balance the objectives of profit maximization and risk minimization. The application of this combined approach results in higher projected annual profits, increasing to 3.4 million and 7 million under different conditions, showcasing the effectiveness of robustness in optimizing strategic decisions. The model offers a more robust, data-driven approach to long-term planning, making it a valuable tool for stakeholders aiming to improve economic outcomes while managing uncertainty effectively.