PFI:BIC - Flexible,equitable, efficient, and effective distribution (FEEED)

该项目开发了一个智能服务系统，以帮助粮食银行等饥饿救济组织向有需要的人灵活、公平、高效和有效地分发粮食。八分之一的美国人在与饥饿作斗争。食品银行试图将多余的食物送到这些需要帮助的家庭，但面临许多挑战。真正的需求是不确定的；捐赠食品的保质期有限；仓库网络和接收机构有能力限制；捐赠的食物供应在时间、数量和质量上都是不可预测的。这些挑战正变得越来越复杂，因为食品银行必须公平地将食品分配给有需要的人，有效地最大化捐赠供应，最大限度地减少浪费，并以具有成本效益的方式分配食品。食物银行使用的现有工具提供信息以促进各级决策，但它们在以下方面的能力有限：(i)提供全面的情况；提出改进建议；(3)从高度动态和不确定的环境中学习并做出反应。feed将综合各种来源的数据，自动预测、可视化，并向决策者学习。制定战略，提高食品收集、分配和资源管理的运营效率，从根本上改变食品银行的运营方式。feed推进了复杂动态分配系统的管理，这些系统需要智能来有效地利用数据来改进决策。本研究项目将涉及探索和发展：(1)一种新颖的随机建模框架，以适应人类决策者动态变化的目标，从而使系统能够随着时间的推移学习并更有效地根据决策者的偏好推荐解决方案；(2)智能操作指导引擎和知识库，帮助食品银行工作人员将复杂的建模操作到日常流程中；(3)人道主义救援领域大数据综合管理框架；(4)促进用户理解和可视化智能系统分析、预测和建议的人机通信模式。本研究通过将捐赠管理和食品分配与供应链管理联系起来，解决了人道主义救济、工程、计算机科学和工业的关键交叉点，并建立了应对这一重大全球挑战所必需的关系。拟议的研究将工业工程和计算机科学的发现与行业最佳实践知识相结合，以开发可应用于全球灵活、公平、高效和有效的人道主义救援行动的技术。

This project develops a smart service system to assist hunger relief organizations, like food banks, in the Flexible, Equitable, Efficient, and Effective Distribution (FEEED) of food to those in need. One in eight people in America struggle with hunger. Food banks try to connect excess food to these families in need but face many challenges. The true need is uncertain; donated food has a limited shelf life; warehouse networks and receiving agencies have capacity constraints; and the donated food supply is unpredictable in timing, quantity, and quality. These challenges are becoming increasingly complex as food banks must distribute food equitably to those in need, efficiently maximize donated supply, minimize waste and distribute food in a cost effective manner. The existing tools used by food banks provide information to facilitate all levels of decision making but are limited in their ability to (i) provide a comprehensive picture; (ii) recommend improvements; and (iii) learn from and respond to the highly dynamic and uncertain environment. FEEED will synthesize data from various sources to automatically predict, visualize, learn from decision maker?s actions, and identify strategies to advance operational effectiveness of food collection, distribution, and resource management and fundamentally transform the way food banks operate.FEEED advances the management of complex dynamic distribution systems that require intelligence to effectively utilize data to improve decision making. This research project will involve the exploration and development of: (1) a novel, stochastic modeling framework to adapt to the dynamically changing objectives of the human decision maker, thus enabling the system to learn over time and more effectively recommend solutions tailored to the decision maker's preferences; (2) an intelligent operational guidance engine and knowledge base that help food bank staff operationalize complex modeling into their daily processes; (3) a framework for synthesizing and managing big data in the humanitarian relief sector; and (4) human/machine communication modalities that facilitate user understanding and visualization of smart system analysis, predictions and recommendations. This research addresses the critical intersection of humanitarian relief, engineering, computer science and industry by linking donations management and food distribution to supply chain management and forges a relationship necessary to address this significant global challenge. The proposed research integrates discovery in industrial engineering and computer science with industry best practice knowledge for the development of technology that can be applied for flexible, equitable, efficient, and effective humanitarian relief operations worldwide.