What is Edge Computing and How Can It Help Manufacturers?

 Machine operator utilizing Lantek WOS to validate and report work completed.

As humans in the 21^st century, we love having information and real-time status updates at our fingertips. Knowing the status of an order you’ve placed seems to be the most gratifying – especially when it’s your dinner pulling up to your doorstep. A critical aspect of the digital transformation for manufacturers is to bring transparency to their operations. This way customers and colleagues can receive status updates for their orders. To become more transparent, manufacturers are beginning to implement new connected machines, sensors, and other tools to report real-time statistics.

More connected devices mean we must transfer, store and access more data. This explosion in connected devices and data aggregation requires a significant amount of network bandwidth. While the global internet infrastructure is constantly updating and offering greater speeds, the number of connected devices will pose an issue to network latency as its bandwidth is currently being pushed to the limits. This is where manufacturers should begin exploring and utilizing edge computing techniques for operational data tracking and real-time reporting. 

What is Edge Computing and how does it work?

Edge computing is when the computation of data happens where, or near where, the data is produced. Data is recorded, analyzed, and accessed locally from connected devices instead of having a remote cloud doing all this work. All the sensors entering our daily operations are continually reporting data. In a cloud only solution this requires the device to collect data and send it to the cloud. The cloud database processes and stores it, then it’s accessible through the cloud again. This requires a lot of bandwidth and high-speed data transmission in order to receive true real-time status.

In a facility utilizing edge computing, each connected device reports back to the company’s on-premise server to record, analyze, and access operational data. This is an internal network that allows for reporting without the constant need for data transmission to and from the cloud. It provides true real-time reporting as all the data points are being collected and viewed locally.

Hybrid Edge-Cloud Computing

You may be thinking, “Our facility just went through the process of transferring all our legacy systems over to cloud solutions. Now I’m being told to bring it back on-premise?!” Firstly, know that your digital transformation efforts aren’t futile and will keep you ahead of the curve. It makes sense to host some systems on the cloud while other systems require too much continual data transmission. For instance, having your customer relationship management, document storage, and other administrative operations on the cloud works just fine. The stream of data isn’t continuous and is generally less urgent.

However, if you’re implementing a manufacturing execution system with real-time workload monitoring utilizing machine sensors and constant reporting, it will generate a considerable amount of data. Since this is operational data, we consider this to be the most important stream of information for day-to-day success. Having a system on-premise allows you to collect and analyze this data knowing it’s truly real-time. This system reports back to your cloud solution in batches – whether it be hourly, daily, weekly, etc.

Think of it like this: edge computing will be necessary for true real-time operational data as more connected devices are introduced to the shop floor. Cloud computing and storage will work for your business’ historical data as a system of record and 10,000-foot view for forecasting.