Manufacturing
The advent of Industrial IoT and new device/sensor technologies is driving improved efficiencies within factories and throughout the supply chain. By feeding data from large amounts of sensors, vast quantities of data can be gathered in just moments.
Smart manufacturing is not just about gathering large data sets through connected devices. At the heart of smart manufacturing is the ability to use that data effectively to improve and make automated decisions, predictions and actions in real-time to optimize industrial output. This requires enormous processing power and it should therefore not be a surprise that smart manufacturing will increase the requirements not just for cloud computing, where the demand for longer-term analytics continue to rise, but also for more processing and storage capabilities at the edge, a trend referred to as edge computing.
While edge is not a new concept in computing, over recent years it has become the key ingredient in the smart manufacturing formula to accelerate digital transformation. Edge computing helps manufacturers turn vast data sets, generated by machines, into insightful and actionable data. It does so by utilizing resources connected to a network, such as temperature sensors, alarms or motor drives. This enables big data analytics to take place at the source of the data.
Edge refers to the OT (Operational Technology) computing infrastructure that resides closest to the sources of data, such as a robotic arm or a conveyor system. These are considered at the ‘edge’ as they tend to exist furthest from the heart of the IT (Information Technology) computing infrastructure, which is typically available in the cloud.
An industrial IoT gateway hosting an open software platform such as IOTech’s Edge Xpert provides a way to connect these devices, facilitate local decision making/actuation and bring the device data to the Internet Protocol (IP) domain for further edge processing or backhaul to the cloud.
These edge platforms need to be able to support standards such as OPC UA, Modbus, CAN bus and other industrial protocols and may also support many wireless protocols such as cellular, WiFi or Low-Power Wide-Area Network (LPWAN) such as Bluetooth, Zigbee or LoRa.
An edge platform provides the following key functions:
- Interoperability - provides the necessary protocol translation for communications to be established between devices that are not able to communicate with each other in a factory.
- Local processing - enables the offloading of computing tasks from smart devices by caching/storing information and acting as a private cloud that can be accessed remotely.
- Quality of service - can maximize the effectiveness of available network bandwidth while minimizing endpoint bottlenecks.
- Security - can be used to support more sophisticated security solutions than those implemented on each individual endpoint, creating a good defensive, in-depth strategy for the whole factory network.
- Local storage - helps save transmission costs by only sending relevant data to the cloud. In many cases it is more efficient to have the edge platform act as the computing node to capture the data and make analytical decisions locally.
Key manufacturing uses cases addressed by the next generation of Industrial IoT and edge computing solutions include:
- Equipment protection and predictive maintenance - a pump with edge computing capability can perform basic analytics to determine if a defined threshold has been exceeded and shut the pump down in milliseconds. Using an edge computing device for such an application means that there is no decision latency and no requirement for Internet connectivity to perform this function.
- Production flow monitoring and optimization - edge computing performs near real-time analytics on multiple data points from sensors in the plant so the data can be processed on a local gateway to provide overall equipment effectiveness (OEE) trends and alerts to operational systems or personnel.
- Supply chain optimization - optimizing supply chain processes for a local facility, factory or an oil field requires data from multiple sources at short intervals to apply optimization algorithms and analytics that will adapt supply-chain plans in business systems such as SCM or ERP. The fundamental capability requires local or factory-level connectivity with decisions made in hours.
Perhaps no industry stands to benefit more from the Industrial IoT than the manufacturing sector. By utilizing edge computing platforms incorporating data storage and computing into industrial equipment, manufacturers can gather data that will allow for better predictive maintenance and energy efficiency, allowing them to reduce costs and energy consumption while maintaining better reliability and productive uptime. Smart manufacturing techniques informed by ongoing data collection and analysis will also help companies to customize production runs to better meet consumer demands.
