All the latest news from the industry weekly compiled by the editorial team for you free of charge.
This eMail is already registered.
An unexpected error occured.
Please accept our Terms of Use.
Registration successful.

CHALLENGES AND SOLUTIONS USING EDGE COMPUTING Edge computing: Driving digital transformation in manufacturing sector

Sep 23, 2022

Data is the epicentre to run the business as it provides valuable business insights and supports real-time control over critical business processes & operations. Since businesses have ocean full of data, the limitation in the virtual world can conspire the efforts to spear ahead. This article illuminates on data challenges through the use of edge computing architecture. - Puneet Walia, Digital Transformation Consultant, Thoucentric

Edge computing has been here for a while, but it is becoming more mainstream with cloud computing, Machine Learning and blockchain. With manufacturing plants having significant processing power on-premises, whether it be in Programmable Logic Controllers, machines themselves or the data centres, this industry is one of the early adopters of edge computing with many potential use cases. Edge computing enables manufacturers to use standard hardware and software to share and access data which is relevant to the manufacturing as well as other associated processes. This further empowers them with having a good governance across the silos.

In the VUCA world, the demand for resilient supply chains is high, plus with pressure on the industry core due to high global competition, there is a need of more flexible and cost-effective ways in how manufacturers run their plants. Driving agile digital transformation with emerging technologies such as edge computing coupled with AI/ML, blockchain or 5G, it empowers manufacturers for securing competitive edge in the ecosystem as well as making them disruption proof.

Where can edge computing be used?

Manufacturing facilities produce large amounts of data, which requires massive processing power, internet bandwidth to drive useful information and make decisions. In addition to these there is requirement of additional paraphernalia such as hardware, security to achieve the desired objectives of digital transformation initiatives. These factors increase costs of technology solutions and make the initiatives unviable, subjecting the investments to unforeseen risks.

Whereas in the case of using edge computing as a technology, the storage systems, intelligence and computing resides close to the component, device or the application. This close proximity significantly reduces the latency as the data is no more required to be sent from the edge of the network to the central processing system and then back to the edge. By processing large amounts of data near the source itself increases overall efficiency of system while decreasing the bandwidth requirement and associated costs. With cost viability, new opportunities for development of more effective applications are created those can be used in remote locations. As there is minimised interaction of edge devices with any public cloud the security is also enhanced. The main edge devices on a manufacturing shop floor are sensors, machines, gateways, laptops, smartphones etc.

Five edge computing use cases for the manufacturing industry

The manufacturing industry is transitioning towards merging information technology with operational technology for more transparency, connectivity, timely data analysis and improved efficiency. In this scenario when the organisations are looking effective ways to modernise their operations, edge computing becomes a priority.

There is an all-time increasing need of reduced emissions, problem detection, downtime minimalisation, creating richer experiences and making supply chains more resilient. Edge computing inspires manufacturers to automate factories through advanced robotics and machine to machine communication closer to source and then take pre-emptive actions such as scanning sheet metal at the right time to detect fatigue failures, monitoring flow through the pipes at any time or keeping track of automated machine cycles. Such analysis helps manufacturers to plan actions in such a manner that failures don’t become obstacles in meeting optimum performance of manufacturing plants.

Here are few of the many use cases those are driving the use of edge computing in manufacturing industry.

  1. Condition-based monitoring

    Edge computing helps the equipment manufacturers to remotely monitor the assets and provides better compliance with SLAs due to precision of data available. In addition to the SLAs, this can also generate new revenue streams by enabling them to provide efficient condition-based maintenance of the assets where customer pays for managed service making uptime a key performance indicator.

  2. Predictive maintenance

    Machine sensor data is not only used to predict failure of the machines but also plan actions in order to reduce the losses associated with unplanned/planned machine stoppages. Accurate prediction of machine failures based on sensor data requires enormous data processing and training of algorithms. To avoid latency and improve the analysis, the algorithms are trained on the central servers and then the copies are deployed for execution at the edge. This concept has potential to be applicable for vision recognition, audio data and local control of robots or other cyber physical systems.

  3. Manufacturing-as-a-service

    Manufacturing site sharing business models are enabled by reducing the site setup times making MAAS an edge computing use case. First, the system needs to be available irrespective of the geographical location of the manufacturing plant while still meeting the stringent latency requirements as it is one of the mission-critical parameters for operating such business model. Second, processing data at the edge overcomes manufacturer’s data security concerns.

  4. AR/VR in the manufacturing plant

    There are lot many AR/VR use cases already deployed on the shop floors. From training the employees on how to use equipment or new processes to; guiding a worker through a hazardous environment or; from assisting the maintenance technicians with remote expertise to; detecting product faults during quality inspections, Augmented/Virtual Reality helps amplifying the human power to perform manual tasks on the factory shop floor. Edge computing helps processing the data on site thus eliminating the problem of latency and make the AR/VR wearables lighter and user-friendly.

  5. Precision monitoring and control

    Edge computing plays a critical role for Artificial Intelligence and Machine Learning. The relevance not only remain to collecting, aggregating and filtering the data to send the outcomes to a central server but also extends to train the Machine Learning algorithms as well as executing them. Given that the amount of enormous data processing required for AI/ ML, a manufacturer may choose to distribute processing across multiple processors, i.e., edges, rather than do this in the cloud.

Edge computing and cloud computing complementing each other

Edge computing and cloud computing are two different technologies which complement each other based upon use cases. Cloud computing will take dominant role when actions require significantly higher computing power, managing massive volumes of data across the manufacturing plants, asset health monitoring, Machine Learning and so on. Use cases that have to do with low latency sensitive information processing, Edge computing would be the best solution as data does not have to traverse over a network to data center or cloud. With that, cloud and edge computing are both necessary to gain most value from varied volume of data based on the use cases & desired outcomes.

Integrating edge computing with 5G

5G was designed to carry massive amounts of data while providing low latency, high reliability and immense bandwidth. However, when millions of devices connect to the network, data from each device starts flowing in, the bandwidth required to transport all becomes tremendous. To achieve the revolutionary latency requirements committed by 5G, the data processing shall be processed at the edge so that it is close to the source or where it is generated.

One use case of 5G and edge computing in manufacturing is the automation of preventive maintenance on machinery and quality sensing of manufacturing lines, which will reduce the amount of human interaction needed. This will free up the resources to instead focus on product development or redefining the manufacturing processes. At the organisational level there are plethora of 5G and edge computing use cases having the potential to touch all roles and functions directly or indirectly.

Edge computing and blockchain

The exponential growth of IoT devices is giving rise to newer and complex solutions those require real time results. Blockchain is a transparent distributed ledger system offering highly secure and reliable transactions and authentications. In Manufacturing industry, blockchain technology provides visibility, security and reliability of the exchanges and transactions of money, parts, materials and products. The major use cases of edge computing and blockchain are lot tracking and defending against counterfeit goods.

Lot tracking – Food manufacturers are required to comply with ‘one up one down’ approach i.e. keeping the track of where the produce come from and where it is going. Similarly, automakers need to track their supplier through to end customers. Blockchain technology gives unique identification to each part, making it possible to pinpoint the problems in their supply chains and take corrective actions at the right time.

Defending against counterfeit goods – Trade of counterfeit products is increasing. Unlike others, using fake, counterfeit and unapproved parts in aircrafts or automobiles can lead to life and death consequences.

In both use cases, achieving real time tracking and identification of goods, materials & spares is a challenge. Waiting for the authenticity checks and reports will lead to additional process turn around times and business loss. To solve this issue, manufacturers and logistics companies have started using blockchain technology where the authenticity can be assured using unique identification generated on blockchain registry which is then coupled with the edge computing enabled wearables are used in the factory shop floors, warehouses, ports for real time validation of authenticity of parts, products and materials.

Conclusion

With various edge computing partners existing in the market, it needs to be ensured that a solution helps in accomplishing following goals:

  1. Manage distribution of software at massive scale

  2. Address network security soncerns, management complexities and limitations of latency and bandwidth

  3. Leverage the edge computing solution that nurtures ability to innovate and handle the diversity of equipment and devices of today’s marketplace

  4. Deploy applications to all edge locations reliably and seamlessly

  5. Maintain openness & flexibility to adopt to evolving needs

Edge computing allows businesses to bring digital world into physical. With a decent pace of adoption, going forward, the edge computing can transform the industry enabling the organisations to explore new business models & work on new opportunities.

Image Gallery

  • Edge computing integrated with blockchain technologies, Chuanwen Luo et al

  • Puneet Walia

    Digital Transformation Consultant

    Thoucentric

Companies related to this article
Related articles