Future-Proof Your Cloud Agility at the IoT Edge
Previously, field-based machines had to be manually configured and monitored, necessitating additional personnel for flexible operation. According to a 2018 report by Gartner Research titled "The Future Shape of Edge Computing: Five Imperatives", there are five imperatives that drive potential edge computing solutions, including latency and determinism, local interactivity, limited autonomy, data and bandwidth, and privacy and security issues. With ongoing demand for real-time insights and localized action, the deployment of edge computing solutions in industrial environments is rapidly increasing. IoT edge devices can deploy cloud-based AI engines at the edge to collect machine data locally and in real time. When combined with Advantech’s OT technology, a variety of industrial machine data can be gathered, processed, and transformed into actionable information directly on the device. By automating data analysis and machine response at the edge, IoT edge computing has significantly reduces the need to transfer massive data between the edge and the cloud, thereby revolutionizing the industrial IoT world.
Configuring AWS IoT Greengrass Core and Azure IoT Edge
Integrated with AWS IoT Greengrass Core and Azure IoT Edge technologies, Advantech Edge intelligence solution can extend cloud intelligence and analytics to industrial-grade platforms to provide various data analysis capabilities, including stream analytics, machine learning, and image recognition. AI-based computing on industrial IoT edge devices enables analytics applications, such as defect inspection and quality assurance, to be performed directly in the field.
WISE-EdgeLink Supports 200+ PLC Devices
Application Scenario: Optimizing Process Manufacturing with Edge Intelligence
Process manufacturing, which is commonly employed in the food & beverage, pharmaceutical, agricultural, and energy industries to mass produce goods, presents different challenges compared to those of mechanical manufacturing. Because process manufacturing relies on the flow of sequential steps, most manufacturing systems utilize steady-state analysis to satisfy all expected uncertain conditions. However, the interdependency of process manufacturing procedures can impact efficiency and quality as insignificant events may be exacerbated into extreme events by the continuous manufacturing procedures. Moreover, numerous sensors and actuators as well as real-time control functions are required to ensure reliable operations and prevent interruptions when extreme events occur.
Advantech’s AWS solution leverages AI and data analysis to match production to demand when the real production model cannot be mathematically constructed. AWS IoT SageMaker is used to build an AI model at the edge via AWS IoT Greengrass in order to facilitate real-time analysis, notifications, and control of sudden uncertain conditions during manufacturing. The inclusion of Advantech’s WISE-Edgelink software, which supports 200+ PLC drivers and communication protocols, enables device data to be collected from diverse sensors and PLCs for transmission between the AWS edge engine, third-party software, and the cloud. This allows all equipment to be monitored and managed remotely via AWS IoT SiteWise.
The Value of Integrating OT and IT with Edge Computing Platforms
To more efficiently realize Industry 4.0 and enhance the reliability of factory/equipment applications, devices should be capable of edge computing. Edge intelligence not only facilitates data-driven decision making, but can also increase overall productivity. In collaboration with diverse software partners, Advantech has developed a new series of edge solution-ready platforms (SRPs) that feature various edge computing capabilities and offer numerous services to satisfy a wide range of market demands.
Advantech's device-to-cloud solution realizes the possibility for edge devices to pass equipment and environmental data to cloud platforms directly. They also enable the utilization of cloud platform tools to perform statistical analysis and visualization of data, so that users can understand the status of the equipment and environment at a glance. In addition to solving typical problems associated with building IoT applications, which tend to be widely distributed and composed of many devices, they also reduce inspection costs, improve immediacy, and facilitate the remote management of decentralized sites.