Edge computing: 5 use cases for manufacturing

This is how a layman – me – explains what production is: it means taking raw materials and transforming them into finished products.

If you want a more formal definition, here’s one from the US Bureau of Labor Statistics: “The manufacturing sector includes factories engaged in the mechanical, physical or chemical transformation of materials, substances or components into new products.”

It feels old-fashioned and highly physical – and perhaps not exactly breeding ground for cyber innovation. However, manufacturing, just like the industrial sector in general, is a natural choice for edge computing and related trends like IoT, AI, and machine learning.

Automation is a big deal in manufacturing and has been for centuries. (The industry even has entire industry publications devoted to the subject.) When the business world at large talks about how people and machines – or people and code, in the case of technologies like RPA and AI / ML – will work together now and in the future, manufacturing CIOs smile and nod knowingly.

There are tons of machines, robotics, sensors, and other devices that generate huge amounts of data. To maximize the value of this data, manufacturing companies need maximum flexibility in their IT infrastructure. For reasons similar to industry, edge computing architecture is not an unlikely choice in production environments – it is natural.

[ Developing an edge strategy? Also read Edge computing: 4 pillars for CIOs and IT leaders. ]

“The manufacturing industry continues to push for cutting-edge applications – from robots that run around the warehouse to sound calibrators to cameras that detect defects on the production line – to advance factory automation and efficiency,” says Brian. Sathianathan, CTO of Iterate.ai. “There is no doubt that edge computing is, and will continue to be, extremely important to the industry. The challenge for CIOs in this industry, however, is how to deploy the power of perimeter systems while ensuring that perimeter applications always remain active and do not destroy their networks. “

Edge computing offers manufacturing CIOs a model for making strategic decisions about what should work, for example, in a warehouse or assembly line, and what should work in a centralized cloud or data center, and what will flow from the cloud at the edge and vice versa.

As Red Hat technology evangelist Gordon Haff recently told us, “The idea is that you often want to centralize if possible, but keep it decentralized if necessary.” And fellow Haff technical evangelist Ishu Verma points out that the perimeter architecture also allows IT leaders to standardize their perimeter operations on the same practices and tools used in their centralized environments.

“This approach allows companies to extend emerging technology best practices to the edge: microservices, GitOps, security, and so on,” says Verma. “This enables the management and operations of perimeter systems using the same processes, tools and resources as centralized sites or the cloud.”

While potentially true for any industry, this is especially important in an industry such as manufacturing, where an organization could very well have thousands of (or more) peripheral nodes running in extremely diverse and challenging environments.

5 examples of edge computing in manufacturing

With that in mind, here are five examples of manufacturing organizations that can use edge computing.

1. Automation of quality control

Again, automation is typically a big deal in manufacturing, although how it manifests itself can vary greatly.

“Manufacturing facilities can have minimal automation up to a fully automated production line,” says Andrew Nelson, Principal Architect at Insight.

Edge / IoT implementations can become increasingly useful as an environment moves to the “fully automated” end of the spectrum.

Edge / IoT implementations can become increasingly useful as an environment moves to the “fully automated” end of the spectrum.

Automating quality control on a production line is a good example, according to Nelson, and is common in contexts such as a canning line in the beverage industry or the packaging process in a food or agri-food environment.

A mix of computer vision, sensors, and other equipment can detect anomalies or other problems; being able to act quickly on that data requires keeping it as close to the process as possible.

2. Warehouse automation

A similar but separate automation use case is in the warehouse, where functions such as inventory management are rich in data and opportunities for greater efficiency.

“Some manufacturers run warehouses near production lines,” says Nelson. “Machine vision can be used to manage inventory levels and help with product picking. RFID / BLE can previously also be used for item positions and quantity levels. Smart shelves can be instrumented with sensors like another data point. “

Sending all of this data to a centralized cloud or data center is probably not the most cost-effective or performance-effective option. Edge implementations create the flexibility to make more optimal decisions about what to do locally in the warehouse, whether for latency, cost, security, or any other reason.

3. Production line diagnostics

We hear a lot about “predictive analytics” these days, but it’s a broad term: its actual value depends on business or industry-specific applications, and manufacturing has a big one: using machine data to monitor and predict with greater accuracy when large numbers of moving parts and pieces in a production environment will break or otherwise require maintenance.

“The [production] the line itself can be instrumented to predict problems with bearings, belts, motors, etc. ” says Nelsons. “In many cases, a line that goes down for maintenance can cost a company a lot. If you can quickly predict or assess problems, you can minimize downtime “and potentially save significant ongoing costs.”

In that context, latency becomes expensive. Local processing of this data can produce tangible financial ROI. And that ROI can be amplified by combining this type of predictive analytics with the QA / QA automation described above by Nelson.

“This can be combined with Q / A processes into a single landscape with multiple benefits and a greater ROI,” says Nelson.

4. Logistics and product monitoring

This category extends the limit, allowing inventory tracking and other uses even as products leave the production environment to other stages in the supply chain.

“RFID and Bluetooth with low emissions [technologies] they can be used to track products as they move through the line and out of production in crates and pallets and also when moving to containers, “says Nelson.” Trucks can be scanned as they enter and exit a warehouse to satisfy both incoming and outgoing product levels “.

It’s a reminder that, as a perimeter server and application, the boundaries of the “edge” can continually expand.

5. The “golden” use case: AI / ML applications

If latency reduction is the most common driver of edge computing strategies, AI / ML workloads are likely to become the golden use case, at least in the manufacturing sector.

“The most powerful production edge deployments depend on the power of the AI ​​powering them, but running intelligent machines smoothly at the edge requires a lot of data,” says Sathianathan, CEO of Iterate.ai.

The problem isn’t a lack of available data – all of the above use cases reflect the reality that manufacturing CIOs are inundated with information. In fact, Sathianathan says manufacturing has an edge over other industries when it comes to AI / ML because so much of an organization’s data is machine-generated.

[ Related read: Edge infrastructure: 7 key facts CIOs should know about security. ]

“Unlike data in other industries which includes a lot more bias and noise, production system data is particularly relevant and valuable ‘gold data’,” he says.

The difficulties arise when trying to send all data from the production site to the cloud or data center. As Sathianathan told us recently, there may be “too much data” to go from a factory or warehouse across the local network and into the cloud and back again.

“That’s not good, because, as production CIOs know, decisions must be made immediately to be effective,” says Sathianathan. “And while some downtime is generally acceptable in standard IT environments, that’s simply not the case in production. The costs of shutting down production lines due to the weakening of perimeter applications can be hundreds of thousands of dollars per minute, there is simply no room for error. “

As edge computing and AI / ML technologies mature, both in terms of infrastructure and in terms of developing lighter applications (via low-code and other tools), they become a perfect match for IT heaven.

“Advances in AI and edge servers with GPU-centric architectures are becoming available, and for production CIOs, it’s a much better solution to start positioning AI applications at the edge,” says Sathianathan.

[ Learn how leaders are embracing enterprise-wide IT automation: Taking the lead on IT Automation. ]

Leave a Comment

%d bloggers like this: