GOALI: A System Theoretical Framework for Modeling, Analysis and Closed-loop Management of Supply Chains of Perishable Products

据估计，地球上每九个人中就有一人遭受饥饿，而供人类消费的粮食中约有三分之一被浪费了。减少易腐食品的浪费是解决世界饥饿问题的关键，因为易腐食品在健康饮食中发挥着至关重要的作用。这些产品衰变相对较快；它们的生存能力从几天到几周不等，这与它们在收获到市场交付供应链上花费的时间相当。因此，对易腐产品进行有效的供应链管理以最大限度地减少浪费是当务之急。这可以通过在运输和储存过程中主动设置环境条件（例如温度、湿度），以及优化运输和交付的时间和数量来实现。该项目将通过考虑产品衰变、产品需求不确定性以及自然生长产品的质量和降解速度的可变性，为易腐产品供应链的最佳管理创造一种新的方法。将开发可扩展到工业供应链网络的计算实现，并使用现有鳄梨供应链的数据进行验证。初步模拟结果表明，食品浪费可能减少15%，冷藏储存的能源消耗最多可减少30%。预计整个供应链成本和食品价格将相应降低，同时二氧化碳排放量也将减少。研究结果有望为减少美国和全球的粮食不安全指明一条明确的道路。这项工作的结果将直接适用于其他时间紧迫产品的供应链管理，如疫苗和移植器官。该项目将支助培训研究生和本科生研究人员，这些研究人员将积极从代表性不足的群体中征聘，并支助为本科生教育编制新的课程材料。本课题将为易腐产品供应链优化管理提供新的系统理论基础。该框架独特地结合了生产计划、反应工程、非线性系统动力学和最优反馈控制等概念，为易腐产品创建了一个急需的供应链管理框架。具体的新颖基础贡献包括，i)作为环境条件函数的产品库存和质量演变的状态空间表示，ii)基于（有限的）实践中可用的测量集的库存和质量状态估计技术，以及iii)可扩展的协作分布式控制策略，并考虑到供应链的结构特性，包括许多，地理分布和经济多样化的实体。这些策略依靠反馈来减轻供应链操作和生物制品衰减率固有的不确定性。该项目的结果将对学者和实践者处理易腐产品供应链管理的方式产生变革性影响，更一般地说，将对供应链和其他大规模存储网络运行中存在的动态和不确定性产生影响。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

An estimated one in nine citizens on Earth suffer from hunger, while about one third of the food produced for human consumption is wasted. Reducing waste of perishable foods is key to addressing world hunger owing to their essential role in a healthy diet. These products decay relatively quickly; their viability ranges from a few days to a few weeks, which is comparable to the time that they spend in the harvesting-to-market delivery supply chain. Effective supply chain management for perishable products to minimize waste therefore is a high priority. This can be accomplished by actively setting environmental conditions (e.g., temperature, humidity) during transportation and storage, and by optimizing the timing and quantity of shipments and deliveries. This project will create a new approach for the optimal management of the supply chain of perishable products by accounting for product decay, product demand uncertainty, and variability in the quality and degradation speed of naturally grown products. Computational implementations that are scalable to the supply chain networks encountered in industry will be developed and validated with data from an existing avocado supply chain. Preliminary simulation results indicate a potential 15% reduction in food waste and up to 30% reduction in energy consumption for refrigerated storage. Commensurate reductions in overall supply chain cost and in the price of food are expected, along with reductions in CO2 emissions. The research results are expected to chart a clear path towards reducing food insecurity in the United States and globally. The results of this work will be directly applicable to the management of supply chains of other time-critical products, such as vaccines and transplant organs. The project will support the training of graduate and undergraduate researchers, who will be actively recruited from underrepresented groups, and the development of new curriculum materials for undergraduate education. This project will lead to a new system-theoretical foundation for the optimal management of perishable-product supply chains. The proposed framework uniquely unites concepts from production planning, reaction engineering and nonlinear system dynamics, and optimal feedback control, to create an urgently needed supply chain management framework for perishable products. Specific novel fundamental contributions include, i) a state space representation of the evolution of product inventory and quality as a function of environmental conditions, ii) state estimation techniques for inventory and quality based on the (limited) set of measurements that are available in practice, and iii) collaborative distributed control strategies that are scalable and account for the structural properties of a supply chain comprising numerous, geographically distributed and economically diverse entities. These strategies rely on feedback to mitigate the uncertainty inherent to supply chain operations and to the decay rates of biological products. The results of this project will have a transformative effect on the way academics and practitioners approach the management of supply chains of perishable products and, more generally, on accounting for the dynamics and uncertainty present in the operation of supply chains and other instances of large-scale storage networks.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.