CNS: Collaborative Research: EAGER: From Computer Networks to Food Networks

由于食品来自全国各地和世界各地，食品供应链是从农场到餐桌的极其复杂的路径。不幸的是，在美国，多达20%-30%的新鲜食品在送到顾客手中时已经变质，同样数量的食物被顾客进一步扔掉，至少部分原因是缺乏新鲜度。因此，了解粮食网络并设计减少损失的机制可能会产生巨大的经济和环境影响。该项目的目的是通过应用和扩展计算机网络科学中的技术来了解和改进食品分销网络，计算机网络可以处理食品网络中看到的规模之外的巨大复杂性。食品网络的独特特性将要求在计算机网络方面进行超越最先进水平的重大创新，因此也将为新颖和新兴的复杂计算机网络提供有价值的结果。“食品网络”的设计和优化带来了许多挑战，需要创新，预计这些创新将使食品网络和计算机网络都受益。具体而言，食品网络涉及独特的“腐烂包”(包装内容物质量的持续下降)、需求驱动和以质量为中心的递送和分发、“包”的网络内捆绑/拆分、多种“有效载荷”的混合以及包源和类型的动态变化。这项提议的目的是应对这些挑战，并为系统地设计和运作粮食网络奠定基础。由于其中许多挑战也与计算机网络有关，因此拟议的工作也将在这方面加以考虑。特别是，研究将审查与网络内分组捆绑/非捆绑、基于内容(而不是基于地址)的路由/分发、多模式数据的服务质量感知转发以及容忍中断操作相关的问题。这项研究还将使粮食网络能够更好地管理，以应对最大限度地利用季节性和当地粮食的新趋势，这就要求粮食来源和分配网络具有极大的灵活性和动态可配置性。

With food sourced from every part of the country and from around the world, food supply chains are extremely complex pathways from farm to the table. Unfortunately, up to 20-30% of the fresh food is spoiled by the time it reaches the customer in US, and a similar amount is further thrown away by the customers at least partly due to lack of freshness. Thus, understanding food networks and devising mechanisms to reduce losses can have huge economic and environmental impact. The purpose of this project is to understand and improve food distribution networks by applying and extending techniques from the science of computer networking, which can handle tremendous amounts of complexity beyond the scales seen in food networks. The unique characteristics of food networks will require significant innovations beyond the state of the art in computer networking and thus will provide valuable results for novel and emerging complex computer networks as well. The design and optimization of a "food network" poses many challenges and requires innovations that are expected to benefit both food networks and computer networks. In particular, food network involve unique 'rotting packets' (continuous decrease in quality of the package contents), a demand-driven and quality-centric delivery and distribution, in-network bundling/unbundling of 'packets', mixing of a large variety of 'payloads', and dynamic changes to packet sources and types. The purpose of this proposal is to address these challenges and lay the foundation for a systematic design and operation of food networks. Since many of these challenges are relevant to computer networks as well, the proposed work will consider them in that context as well. In particular, the research will examine issues relating to in-network bundling/unbundling of packets, content based (rather than address based) routing/distribution, QoS aware forwarding of multi-modal data, and disruption tolerant operation. The research will also allow for better management of the food networks in the face of emerging trends in maximal utilization of seasonal and local foods which requires significant flexibility and dynamic configurability in the food sourcing and distribution networks.