Provisioned Data Distribution for Intelligent Manufacturing via Fog Computing
Provisioned Data Distribution for Intelligent Manufacturing via Fog Computing
复制标题
DOI：
10.1016/j.promfg.2019.06.158
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Riddhiman Sherlekar;B. Starly;P. Cohen
中科院分区：
文献类型：
--
作者：
Riddhiman Sherlekar;B. Starly;P. Cohen
从简单的设备到工厂车间的复杂机器，再到企业级的互联网，“事物”的数量正在呈指数级增长。这种联系也导致产生了大量的数据，导致“数据”现在被认为是更广泛的制造业的核心资产之一。然而，这一资产的可用性几乎没有利用中小型制造企业（SME）-美国的“Mittelstand”。中小企业如何共享某些类型的数据，同时又能够保留对自己数据的所有权和控制权？中小型企业如何利用这些不同形式的数据计算，为其客户和自身带来利益？在本文中，我们提出了一种分散的数据分发架构，以民主化制造业使用雾计算范式产生的大量数据的潜在可用性。该架构利用了云制造的行业可扩展中间件扩展，可以安全地过滤并将数据从车间的物联网制造机器传输到云上的潜在用户。这项工作还展示了一种以数据为中心的方法，该方法允许在雾层内横向共享对等数据，以服务于云用户。我们通过涉及各种类型的制造数据的案例研究证明了雾中间件基础设施的可行性。
The number of ‘things’ ranging from simple devices to complex machines on the factory floor connected at the enterprise level and to the broader internet is growing exponentially. This connection also leads to a tremendous amount of data generated leading to ‘Data’ now considered one of the core assets in the broader manufacturing industry. However, the availability of this asset is hardly made use of by Small and Medium scale manufacturing enterprises (SME) - the ‘Mittelstand’ of America. How can certain types of data be shared by SME companies, yet have the ability to retain ownership and control over their own data? How does SME leverage computing on these diverse forms of data for the benefit of its clients and itself? In this paper, we propose a decentralized data distribution architecture to democratize the potential availability of large amounts of data generated by the manufacturing industry using the Fog Computing paradigm. The architecture leverages an Industry scalable middleware extension of Cloud manufacturing that securely filters and transmits data from IoT enabled manufacturing machines on the shop floor to potential users over the cloud. This work also demonstrates a data-centric approach which allows peer-to-peer data sharing laterally within the fog layer to serve cloud users. We demonstrate the feasibility of the Fog middleware infrastructure through case studies that involves various types of manufacturing data.