Scalable Enterprise Systems: A Modular, Scalable Approach to Modeling and Analysis of Semiconductor Manufacturing Supply Chains

半导体制造供应链非常复杂，跨越地球仪各地的多个制造地点。这笔赠款提供的资金将允许多学科研究团队利用在工厂控制中获得的经验来开发半导体供应链的控制制度。目标是高效且有效地a）对物理实体进行建模和仿真，B）对决策算法的范围进行建模和仿真，以及c）将两者进行接口。研究小组已经为此迈出了第一步。为了描述物理系统，开发了一个基本模块，可用于表示工厂，运输链路或仓库。基本模块由容量子模块和时延子模块组成。供应链通过将基本模块串联或并联连接在一起来建模。这种方法提供了合理的执行时间，同时仍然捕获这些系统的定性行为。因此，研究团队将首先确定合理的方法来参数化基本模块。接下来，将研究支持半导体供应链建模和分析并确保可扩展性的软件架构。真实的半导体供应链问题（由工业参与者提供）将用于解决参数化和架构问题。最后，将研究如何在下一代ERP系统中实施该方法。如果成功，这项研究将通过确定如何在这些系统中包括可扩展的“假设”功能，对下一代ERP系统的系统架构产生影响。它还将通过允许一组人描述其工厂的工作方式，另一组人指定企业决策策略，促进可扩展企业系统中的协作决策。最后，它将允许通过提供新的建模和分析功能来分析各种供应链设计问题。

The semiconductor manufacturing supply chain is very complex spanning multiple manufacturing sites in various locations around the globe. This grant provides funding that will allow the multidisciplinary research team to use experience gained in factory control to develop control regimes for semiconductor supply chains. The goal is to efficiently and effectively a) model and simulate the physical entities, b) model and simulate the range of decision algorithms, and c) interface the two. The research team has taken a first step toward this. To depict the physical system a basic module was developed that could be used to represent a factory, a transportation link, or a warehouse. The basic module is made up of capacity and delay sub-modules. A supply chain is modeled by connecting basic modules together in series or parallel. This approach provides reasonable execution times while still capturing the qualitative behavior of these systems. Therefore, the research team will first determine reasonable ways to parameterize the basic module. Next, software architectures to support the modeling and analysis of semiconductor supply chains and ensure scalability will be investigated. Real semiconductor supply chain problems (supplied by the industrial participants) will be used in addressing both the parameterization and architecture issues. Finally, how the approach can be implemented in next-generation ERP systems will be studied.If successful, this research will have an impact on the System Architecture of next-generation ERP systems by determining how to include scalable "what-if" capabilities in these systems. It will also facilitate Collaborative Decision Making in scalable enterprise systems by allowing one group of people to describe the way their factories work and another group of people to specify enterprise decision policies. Finally, it will allow for the analysis of various Supply Chain Design issues by providing new modeling and analysis capabilities.