EAGER: Cybermanufacturing: Design of an Agile and Smart Manufacturing Exchange: Enabling Small Businesses through Standardized Protocols and Distributed Optimization

这一早期概念探索性研究(AGERGE)奖支持对敏捷制造交换系统(MES)设计的基础研究，在该系统中，原材料供应商、装配商、运输公司等将通过标准化协议参与，以完成复杂的制造订单。这一设计将为智能软件中介层(即“中介”)奠定基础，使MES能够自我学习并适应动态/多样化的服务请求和资源可用性，并在复杂的信息生态系统中支持服务提供商和用户的大型网络。美国的经济竞争力依赖于制造业智能大规模定制系统的新方法和创新方法，这些方法将使需要几乎即时履行的小规模和多样化订单的处理成为可能。MES将通过支持按需整合资源、从容不迫地从故障中恢复以及在不中断操作的情况下进行动态调整来实现这一转变。为了达到这些目标，研究将集中于通过实时动态地进化优化策略来适应新出现的系统行为。用户和供应商将连接在一个动态的制造网络中，该网络将容纳多个产品流、供应商和他们之间的链接的不确定性以及容错能力，以便在网络组件发生故障的情况下提供服务。这种水平的适应、无缝的效率和不间断的服务将构成迈向智能MES的重要一步。研究目标将通过设计分布式实时优化和知识发现框架来实现，该框架将解决用户、经纪人和供应商组成的动态制造网络中的工作流优化、资源分配和数据驱动的性能预测。具体的研究任务包括在线准入控制策略、动态生产计划、用于预测的服务级别性能的分析和预测、不确定情况下的分布式动态资源分配方法以及支持人类决策者和态势感知的可视化分析技术。

This EArly-concept Grant for Exploratory Research (EAGER) award supports fundamental research on the design of an agile manufacturing exchange system (MES) in which suppliers of raw materials, assemblers, transportation companies, etc., will participate through standardized protocols to fulfill complex manufacturing orders. This design will provide the foundation for a smart software mediation layer (i.e., a "broker") that will enable a MES to be self-learning and adaptive to dynamic/diverse service requests and resource availability, as well as support a large network of service providers and users within a complex information ecosystem. The economic competitiveness of the U.S. dependss on new and innovative methods for intelligent mass customization systems for the manufacturing sector, which will enable the processing of small-sized and diverse orders that demand almost instant fulfillment. The MES will enable this transformation by supporting on-demand integration of resources, graceful recovery from failures, and dynamic adaptation without any disruption in operations. In order to meet these goals, research will be focused on adaptation to emerging system behaviors by dynamically evolving optimization strategies in real-time. Users and providers will be connected in a dynamic manufacturing network that will accommodate multiple product flows, uncertainty in links between providers and themselves, and fault tolerance to provide service despite failed network components. This level of adaptation, seamless efficiency, and uninterrupted service will constitute a significant step forward towards a smart MES. The research goals will be accomplished through the design of a distributed real-time optimization and knowledge discovery framework that will address workflow optimization, resource allocation, and data-driven performance prediction in a dynamic manufacturing network of users, brokers, and providers. The specific research tasks include online admission control policies, dynamic production planning, analysis and prediction of service-level performance for forecasting, distributed methods for dynamic resource allocation under uncertainty, and visual analytics techniques to support human decision makers and situational awareness.