Edge, Fog and Cloud computing clarified

The digitisation of manufacturing will have a huge impact on organisational growth and profitability in ways not yet clear. More data offers a better understanding about production process and gives users the opportunity to link different events. For example, why something is happening may result from changes directly not associated with a process. It is also the platform for delivering advances in preventative maintence to reduce unplanned downtime. Converting the data delivered by IIoT and Industry 4.0 needs additional computing and data handling capabilities.To achieve this, producers need ways of handling with the massive volumes of data that are available to them. This includes edge, fog and cloud computing, and the adoption of artificial intelligence tools. Tim Foreman of Omron Automation thinks it is time to clarify edge, fog and cloud computing in relation to artificial intelligence, and offers an analogy to industrial manufacturing using the body’s nervous system.

Edge, Fog and Cloud computing

Edge: Spinal reflex AI

You can compare data mining at the Edge to a spinal reflex. Monitoring of lines and devices with real-time sensors and data on the machine level takes happens in in microseconds. This removes processing latency as data is not sent from the edge to a central processing system and then back. Users can check a machine condition real-time, but where the data volumes are small. Real-time data processing at the Edge enables an immediate response to an abnormal situation in a process. With AI at the Edge, manufacturers can control complexity and security.

Technology offering seamless integration to the Cloud from the machine controller with embedded secured IoT protocols is also required. To translate information into action, manufacturers need efficient control and monitoring for a more natural, proactive relationship between operator and machine.

Typical handling of this is by PLCs or industrialised PCs without any data needing to go up into the local Fog or remote Cloud based IT infrastructure. However, Omron gives users the option to connect their machine to Fog level SCADA software, in-house custom software solutions or long-term historian databases.

Fog and Cloud: Cerebral AI

Fog computing works to bridge edge/OT with enterprise IT systems. It puts intelligence at the local area network (LAN) level.

Fog computing includes the data processing, storage and networking technology that sits beneath the Cloud. Big Data mining in the Cloud, known as ‘Cerebral AI’ allows manufacturers to analyse data, collected from an entire factory over several years.

However, communication between the manufacturing floor and the Cloud requires seconds to minutes of processing time. This also requires open and secure standards such as the MQTT protocol and the OPC UA communications standard for safe and easy transformation of machine and system data into high value information. Reacting to tiny process deviations in microseconds is impossible for AI in the Cloud. Omron also offers robust Industrial PCs residing on local Fog level networks. These will run SCADA software, or other Windows/Linux based applications for visualisation or batch data analysis

Implementation survey

According to US survey by Automation World, manufacturers are mapping out a route that covers fog and cloud. They address new predictive maintenance and performance monitoring applications at the same time. Almost half of the respondents already have edge computing underway and rather more have cloud. 20 percent cited fog computing, considered a superset of edge computing used to bridge OT with enterprise IT starting to take root.

Fog and edge are enabling technologies and standards that give IoT users and technology providers with more options. Removing the limits of centralised cloud servers means IIoT is much more distributed and flexible in the services providers can offer.

 

Table from a survey by Automation World Journal