Could Edge Computing Transform Manufacturing?

2 Aug , 2019  

Edge Computing may seem like a new buzz-phrase, but how does it work and is it set to be the computing paradigm to move us beyond well-established cloud architectures? Could it even replace the cloud?

Essentially, Edge Computing seeks to move computation away from an all-controlling data centre and distribute it to the edge of the network where it can access other computing systems – particularly data storage – and perform tasks on behalf of the cloud. This improves data transmission speeds and reduces bandwidth usage: a highly desirable for industrial manufacture which is set to become more autonomous and responsive as a result.

The growing number of Internet of Things (IoT) devices now in an industrial setting (Industrial Internet of Things [IIoT]), has enormously heightened pressure on data centres and network bandwidths. The consequence is slower transfer rates and response times – critical requirements to many industrial applications.

How can IoT Edge improve manufacturing?

To overcome the problem, it is necessary to relieve strain on the data centre. This can be done by allowing other information in from smart devices such as telephones and sensors: Edge Computing can store and manage the data (known as caching), and serve it for further use and at a faster speed when next required. As ever smaller devices are packed with more and more analytical capabilities, Edge Computing is in pole position to make the most of the opportunities it offers.

Cloud infrastructures are now integral to businesses across the world, and the offerings of the big three providers – Amazon, Microsoft and Google – dominate the market. The other major player is IBM, which provides the tools used to write cloud-computing applications, and as such is the middleman in the cloud wars.

All of this means that there is little room for performance growth in the cloud industry, with most of the new opportunities lying at the edge. And although cloud computing is often touted as being a huge cost-saving solution, it is often not the case. High-capacity cloud servers with high-volume data transfer capabilities come at considerable set-up, maintenance and subscription costs – and business is naturally cautious about such outlay.

Where cloud has already reached its limitations, Edge Computing plays an important role, as evidenced in autonomous vehicles. GE Digital ( estimates that for every eight hours of driving, autonomous cars generate around 40TB of data, and claim sending such a volume to the cloud would be ‘unsafe, unnecessary, and impractical.’ Edge Computing can untangle the information and use it selectively to allow the fastest possible vehicle functions while safeguarding passengers and pedestrians alike.

While Edge Computing is not new, its viability has been greatly enhanced by the ability of ever-smaller technical devices containing ever-larger computational capacity, making the data relatively inexpensively to access.

Use cases in which Edge Computing will become vital for industry may include fieldwork in remote locations with low and/or intermittent connectivity; situations where access and rapid analysis of material data is required for real-time analytics (e.g. technicians out on-site checking hardware performance); and being able to work with lower bandwidths by minimising network latency; i.e. the time it takes for information to travel to a data centre, be processed, and return to the endpoints.

Agile in manufacturing

This change in network architecture makes manufacturing more agile in core day-to-day functions. All the processing components essential for operating a smart manufacturing facility are available onsite, and can be distributed across the supply chain, making connectivity to a central data centre is less of a necessity.

Decentralising control in this way means that a malfunction need not stop the whole manufacturing process because it can be repaired in isolation to the whole operation. Hardware failure and cyber-attacks also become less of a threat due to the non-centralised nature of the network. Incorporation of the cloud has allowed the industry to gain a greater understanding of the data at its disposal, and they have used it to good affect – reducing energy costs and increasing output.

The cloud is still crucial to the IIoT, but it has limitations when processing huge volumes of data – and this is where Edge Computing comes into its own by helping ensure peak manufacturing performance is maintained. And when the IIoT detects faults in an automated system, Edge Computing can ensure a rapid response.


The principle of Edge technology represents a wider trend away from centralised control and is part of a broader discussion about centralised versus distributed systems. Here, as in other areas, engineers are discovering that structures can be more predictable and sturdy when not constructed around one critical location.

According to the Vodafone IoT Barometer 2018, the proportion of companies using Edge Computing has more than doubled, and nearly all using the technology report a return on investment (an average of 19%).

Looking at the whole gambit of business types in the private and public sector, the report provides sound evidence of the growing need for businesses to lessen reliance on the cloud and take advantage of the benefits available from the new Edge paradigm. As the use of IoT and IIoT becomes a new standard in manufacturing, Edge Computing can give a competitive edge – a consideration above all others that will secure its place in the factory floor.

, ,

Leave a Reply

Your email address will not be published. Required fields are marked *

Automation Engineer

Automation Engineer