Home IIoT Edge Computing is the Key to the Evolution of IoT – Introducing Edge Computing Use Cases

Edge Computing is the Key to the Evolution of IoT – Introducing Edge Computing Use Cases

Edge Computing is said to hold the key to IoT evolution. Why? In this blog we’ll introduce Edge Computing, explain the connection between IoT and Edge Computing, and share examples of Edge Computing use cases.

What is Edge Computing

In computer networks, the cloud – for example, a data center – is located at the center of a network. But what does the term “edge” represent? The edge could be a smartphone or tablet, or industrial equipment that plays an active role on the front lines of the field. Edge Computing processes data at or near to the “edge” or operations.
With that being said, Edge Computing does not limit data processing just at the edge – it is a distributed open IT architecture known for having strong distributed processing power.
While improving response speeds by processing data on the edge side, data that does not require high speed is accumulated on the cloud side. By performing distributed processing in this way, Edge Computing maximizes the value of data and optimizes the utilization method required for each data.
For a more detailed overview of Edge Computing and differences from cloud and on-premises, please read this blog:
Why is Edge Computing attracting attention – What are the differences compared to the cloud and on-premises | Stratus Blog

Edge Computing adds precision to IoT

Edge computing is said to be an indispensable technology for pushing IoT to the next stage.
What impact will Edge Computing have on IoT, and how will IoT change with Edge Computing?

Improved real-time information

The most obvious impact of Edge Computing is the improvement of real-time information and data. Oftentimes when data is sent and received in the cloud to be processed, there is a time lag – between several hundred milliseconds to several seconds – that may occur. Even as sensors become sophisticated and send more accurate data, a time lag in data processing can impact equipment, slowing down and impacting operations. Edge Computing reduces time lag by processing the required data at the edge, enabling real-time responses.

Traffic optimization

With the spread of IoT, a huge amount of data is being sent and received, and the amount of data will continue to expand in the future. Aggregating all this in the cloud increases communication traffic and causes data congestion. By distributing processing at the edge, organizations can reduce both data congestion and communication costs.

Strengthening data security

Another benefit of Edge Computing is enhanced data security. Processing data at the edge of operations reduces the amount of communication with the outside, protecting this data from being exposed to cyberattacks or data breaches.

Establishment of a business continuity plan

Edge Computing can also be an effective tool for business continuity planning (BCP). If the necessary data is handled on the edge side, operations can continue even if the cloud server goes down. As mentioned above, the most important feature of Edge Computing is the distributed processing capability. Distributing data leads to distributing risks and increasing business continuity.

Taking IoT to the next level

By connecting Edge Computing and IoT, organizations will see the above benefits. This will help companies make accurate management decisions and improve business continuity.

Cooperation with the cloud increases the value of Edge Computing

Oftentimes, cloud computing and Edge Computing are compared. But in reality, Edge Computing and cloud computing are not mutually exclusive.
Let’s take a look at why and how the cloud and edge work together.

Separation of cloud and Edge Computing

Initially, Edge Computing was considered a technology that just improved processing speeds by dividing processing between the edge and the cloud.
However, with the evolution of IoT, it has become necessary to process large amounts of data at high speed near IoT devices. Because of this, there has been a division of the two roles – data processing and data accumulation. Processing functions remain mainly at the edge, and analysis and machine learning functions, for example, are now done in the cloud. It’s important to understand how the cloud and the edge work together, as the ways they are leveraged will depend on how each is linked.
In the future, Edge Computing is expected to have the capabilities to generate analysis and machine learning algorithms, just not quite yet. Currently, the cloud side oversees “big-picture analysis and large-scale judgment”, and the edge side performs “local analysis and instantaneous judgment”. It is important to take advantage of the both the cloud and edge capabilities.

Add the benefits of the edge to the benefits of the cloud

There are advantages to operating in the cloud, including a high degree of freedom, scalability, and flexibility. These features of cloud operations are not limited. Edge Computing combines both processing at the edge and in the cloud – users can take advantage of real-time processing at the edge as well as the benefits of operating in the cloud.
By leveraging both edge and cloud, there is also a reduced risk of network and latency issues. This is because an Edge Computing platform processes essential, local data at the edge and sends data that is not time-sensitive to the cloud.
Edge Computing also protects confidential data by keeping it in-network without exposing it to threats when sending to the cloud.
This combination of the edge and the cloud has a positive effect on business operations and management.

Use Cases of Edge Computing

Below we outline several use cases of Edge Computing as well as the benefits. Here are some examples.

Environmental sensing that detects failures by sound

Experienced engineers may hear machines in operation and detect the presence of signs of failure and the need for maintenance. Although it is difficult to express it as an accurate index, there is one aspect that Japanese manufacturing has been supported by such “on-site perception”.
It is a challenge for Japan’s manufacturing industry to pass on the maintenance know-how related to facilities and equipment, which has become dependent on skilled engineers, through digitization.
Solutions have been developed to detect abnormal noises and notify employees of signs of failure. These operation sounds are collected by a microphone and the system then calculates the degree of the anomaly. This data is aggregated and accumulated in the cloud, used for big data analysis and machine learning, and sent to a remote monitoring department.
Edge Computing is a tool used for processing and collecting operating sounds and calculating the degree of anomaly. Anomalies can be detected on the edge side, enabling advanced processing with low latency.

Real-time weather forecast simulation

The distributed processing power of Edge Computing is leveraged for real-time processing of observation data. This data varies by region.
To improve the accuracy of weather forecasting at a certain point, it is necessary to subdivide the forecasting range. However, the more segmentation progresses, the greater the amount of data sent to the data center. There are limits to the communication volume in terms of cost, speed, and processing capacity, and as a result, there are limits to the segmentation of the forecast range.
But with the introduction of Edge Computing, organizations can go beyond that limit. By performing processing at the edge for each subdivided range and sending the data necessary for wide-area prediction to the cloud, it becomes possible to make the necessary predictions where they are needed.
The day may soon come when Edge Computing will enable highly accurate, real-time weather forecasts at specific locations.

Making agriculture more efficient with drones and robotics

Edge Computing is also expanding its potential for use in the agricultural sector.
In the past, fruit harvesting required people to visit the site and make rounds to visually determine whether it was time to harvest. Now, drones and robots perform this work and a system has been put into practical use.
Equipped with image recognition using a camera and AI, AI identifies the acquired image using Edge Computing, analyzes the degree of maturity from the color and shape of the fruit, and detects pest diseases from the color and shape of the leaves. In addition, it is also possible to visualize the harvestable number within the activity range and predict the optimal harvest time.
Edge Computing is applied to this processing, and smooth operation of the robot is achieved by separating image identification at the edge and data analysis in the cloud.

Optimize store operations with staff behavior analysis

In small-scale stores, such as convenience stores, one staff member must be in charge of multiple tasks. Some of these tasks could include product stocking, inspection, cash register work, and in some cases cooking work. Because of this, it is necessary to move frequently without staying in one place in the store.
In order to improve the efficiency of staff movement in the store, solutions have been developed that analyze staff behavior and optimize store placement. Images from cameras placed inside the store are sent to a computer inside the store, where the person is identified and photographed at the edge. As a result, data on how staff move is accumulated in the cloud, and data analysis is used to calculate in-store layouts that achieve optimal flow lines.
This technology can be applied not only to staff, but also to customer purchasing behavior, such as what kind of product display will promote purchasing behavior. However, the challenge is how to ensure that the extracted personal data does not contain elements that identify individuals.

A smart factory that controls the operating status in real time

While systems using IoT and Edge Computing are being developed in many fields, the field of manufacturing, especially the smart factory, continues to evolve the most rapidly.
A smart factory is a factory where all devices and sensors are connected to a network. This not only visualizes the operating status, but also connects and links all kinds of information, such as forecasts and corrections in terms of production management, inventory management, and incoming/outgoing logistics management.
In addition, some factories have introduced a system to prevent human error by sensing the movement of people on the production line with video as well as checking for abnormal behavior.
Edge Computing has become an indispensable technology for smart factories that need to process huge amounts of data and respond in real time.

Medical equipment system in a large-scale hospital that supports community medicine

Edge Computing has also been introduced in the medical field.
Large hospitals, which play a central role in regional advanced medical care, use a large number of medical devices. The utilization of the data output from these devices has been an important issue. Because of this, medical device comprehensive solutions that integrate medical devices and hospital systems are being developed to coordinate and streamline operations, manage device data, and advance the medical care provided.
This solution enables efficiencies and data utilization by linking medical devices and hospital systems. As the system is deeply involved in medical operations, it cannot be stopped. It also handles sensitive data related to patient care, which requires a high level of security.
Edge Computing was combined with this system to increase availability and safety.
There is a connection between advanced medical care and data utilization, such as filtering data from medical equipment and transferring it to electronic medical records and departmental systems.

Meteorological Observation Systems Supporting Safe Flight, Departure, and Landing of Aircraft

Weather observation systems for aircraft require real-time information and stable operations in order to ensure the safe flight of aircraft traveling at speeds of hundreds of kilometers per hour. It is a system that cannot tolerate outages or delays.
Considering the real-time nature of information and continuity of operation, Edge Computing is considered the best choice.
It is operated as a system that constantly collects meteorological information such as wind direction and speed, visibility, cloud base altitude, temperature, humidity, rainfall, and atmospheric pressure, and collectively processes and displays this observation data.

Marine Oil Transportation Management System at Energy Transfer Terminal

At integrated energy distribution bases that support urban areas and large-scale industrial areas, it is important that gateway servers for marine oil transportation management operate 24 hours a day with no unplanned downtime.
In terms of operational continuity, there is an advantage in performing distributed processing rather than building a large-scale server within the facility.
On the other hand, it is not uncommon for plant facilities to not have an environment suitable for installing servers, or to lack human resources for operation.
Maintenance must be performed with a small number of personnel and that the equipment can never stop operating.
Under these conditions, Edge Computing, which can ensure efficient data processing and stable operations, is being used in plant facility systems.

Edge Computing: Increasingly Important in an Era Where More Accuracy is Required

As we saw in this article, Edge Computing is used in various places in conjunction with IoT. At manufacturing sites, IoT has enabled the visualization of various things, but the use of IoT is progressing to the next stage. That is the second stage of IoT utilization, “how to utilize the collected data.” Currently, Edge Computing is required to perform highly accurate processing without impairing the real-time nature of data. Edge Computing will become an indispensable technology for IoT utilization in the future.

Related Posts