Development and Solution of a Multi-Stage Planning Model Based on a Greedy Optimization Algorithm Under Multiple Constraints

This paper presents a comprehensive approach to establishing a multi-stage planning model under complex, multi-constraint conditions. The model is designed to optimize decision-making over a seven-year period, with the primary goal of maximizing overall fitness. At each stage, specific conditions from the previous period are analyzed to inform and guide production decisions for the next stage. Recognizing the diverse origins and characteristics of various products, we conduct a secondary categorization to address these differences effectively. Additionally, we incorporate overproduction penalties under two distinct scenarios to reflect realistic production challenges and constraints. To solve the intricate mathematical model, we employ a greedy algorithm. The application of this algorithm demonstrates its capability to handle the model’s complexity, producing solutions that maintain high fitness levels across all stages. The results confirm that the greedy algorithm not only efficiently solves the model but also adapts well to the dynamic conditions of each stage, ensuring optimal decision-making throughout the entire planning period.