Agrifood system actors operate within diverse sociocultural, economic, and biophysical settings. For growers, crop planning, usually a yearly business plan, is a key decision to make on when, what, and how many to plant. It is a challenging task as it deals with multiple constraints in volatile economic and/or climate environment. Most crop planning models have difficulty in adapting to changing situation. In this study, a parallel system of crop planning composed of the artificial system, computational experiment, and parallel execution is proposed. The farmers are described as agents, and the decision is made based on the heuristic searching of optimal plan; the adaption of plan is triggered autonomously given strong environment changes. Focus is given to economic environment, which is indicated as product price. In a case study, the economic environment of the artificial system is built based on the monthly and weekly price information for 13 products during 7 years. The computational experiment provides the initial cropping plan and harvest time, with social and ecological constraints. Result shows that the cropping plan can further adapt to price variation. This flexible cropping plan system can strengthen the capability of cooperatives serving small-scale farmers.

中文翻译：

---

基于价格预测的平行作物计划

农产品系统参与者在不同的社会文化、经济和生物物理环境中运作。对于种植者来说，作物计划（通常是年度商业计划）是决定何时、种植什么以及种植多少的关键决定。这是一项具有挑战性的任务，因为它要处理多变的经济和/或气候环境中的多重限制。大多数作物计划模型难以适应不断变化的情况。在这项研究中，提出了一种由人工系统、计算实验和并行执行组成的作物计划并行系统。农民被描述为代理人，基于最优方案的启发式搜索做出决策；考虑到强烈的环境变化，计划的调整会自动触发。重点关注经济环境，以产品价格表示。在一个案例研究中，人工系统的经济环境是基于 7 年内 13 种产品的月度和周度价格信息构建的。计算实验提供了具有社会和生态约束的初始种植计划和收获时间。结果表明，种植计划可以进一步适应价格变化。这种灵活的种植计划制度可以增强合作社服务小农的能力。