Ground your understanding of Digital Twins through real manufacturing use cases and customer stories featuring top manufacturers globally. Learn why top manufacturers are investing in Digital Twins for operational excellence.
Edge computing indicates that data is reckoned close to the source.
Edge analytics is a process where a layer of intelligence—such as visualization or alerting—is added to Edge data.
Edge analytics solutions for manufacturing are transitioning from “nice to have” to “vital.” Amidst all the buzz around Edge technology, though, many manufacturers have key questions about Edge analytics:
This introduction to Edge analytics in manufacturing aims to address these questions and more.
Edge analytics is the process of running live production data through analytics algorithms, Artificial Intelligence technologies, business applications, or other devices as soon as it comes in from IoT devices. With Edge analytics, data is collected and analyzed as close to the source as possible instead of sending data to a cloud or data center for later analysis.
All Edge technologies exist along a continuum. Essentially, anything that happens between the place where data is generated (e.g. at the IoT device computing data from a machine or process) and the Cloud (where data is stored) is part of the Edge continuum.
At any point along the Edge continuum, data can be accessed or utilized by people, technologies, and systems.
Edge data is less prone to latency issues and lag times since it is closer to the data creation point. The data used within Edge analytics technologies are closer to the place where data is being generated. For this reason, data used in Edge technologies are considered “live” data.
As previously stated, Edge analytics refers to the usage of algorithms or autonomous technologies running on data close to where data is generated. After going through these algorithms or devices, data is made immediately available to users via a dashboard, notification, or visual interface that easily illustrates what is happening.
Edge analytics refers to the usage of algorithms or autonomous technologies running on data close to where data is generated.
For example, Edge analytics technologies can identify whether the current production volume is running at, below, or above historical production levels. It could track good vs. bad parts and send an alert if out of spec based on real-time conditions.
Edge analytics tools can also monitor and display a machine’s status in real-time, helping teams stay notified about real-time performance. Aggregating the status of a single or fleet of every machine in your facility can give you a 360–degree view of current machine performance without ever stepping foot onto the factory floor.
Data generation in manufacturing is surpassing network capacities. For manufacturers producing and collecting high volumes of data at rapid speeds—say, in the case of carbonated beverage producers or CPG companies—there is likely more data than teams know what to do with. It’s simply not possible to store every piece of data for future analysis. Not only can storage be costly, but it can also increase the risk of cyberattacks.
However, it may also not be necessary to store every piece of data for future analysis. For example, you would want to know if a defect has occurred in production to prevent it from impacting the entire batch. With Edge analytics, teams can receive alerts on these types of events to reduce and minimize problems in production from escalating.
Tools like the Snapshot App from Braincube trigger data from a set period of time after an event like a defect or machine outage. This allows you to extract value from data when it’s the most timely or relevant. The data is captured so you can analyze it deeper when warranted.
Edge analytics technologies enable operations to see what’s happening right now. By making valuable production insights available in live time, teams can not only react faster: they can also react more efficiently. Properly identifying a problem while it’s occurring helps ensure that processes are running at optimal conditions at all times.
Eventually, manufacturers will strive towards even more autonomous technologies. Edge technologies are easy-to-implement solutions that save teams the hassle of waiting for data to be in the cloud, slowing down their reaction times.
By making valuable production insights available in live time, teams can not only react faster: they can also react more efficiently.
For example, a high-volume beverage manufacturer may track the total number of bottles filled at the end of the day. However, it would be more meaningful to know the average good vs bad parts per minute or hour. Teams may also prefer to know that production has dropped during the middle of a shift so that they can quickly adjust to meet quotas.
Giving teams tools that focus their efforts goes beyond simply saving time and resources. Edge technologies can better connect workers to their roles and their impact on production. Tools like Braincube’s Edge analytics applications give teams the ability to visualize, troubleshoot, and pinpoint what’s happening, enabling them to be more efficient and making them feel connected to their work.
Edge applications also help companies up-skill workers and combat labor challenges. With the rise of the Citizen Data Scientist, employees across the organization are expected to bring additive value. Without coding skill requirements or the need for complex manual analyses, these ready-to-use applications put valuable information right into the hands of the operators, maintenance teams, and other individuals who can directly improve production.
Edge analytics technologies put meaningful live insights in front of operations teams. Teams can see when production levels drift, when a machine goes down, or if uptime levels are below historical averages. Seeing these changes and notifications in live time empowers employees to react quicker to events that could negatively impact production.
Alerting and other condition monitoring apps are foundational in a data-driven culture. Tracking vibrations, triggering automated alerts, or building highly visual dashboards provide quick wins with lasting value to manufacturers.
By their very nature, Edge solutions are decentralized data systems because data is gathered and processed close to where it is produced. This makes Edge technologies easy to scale because they don’t rely on a centralized data system. Edge technologies are also great for environments with low bandwidth, which is another scalable benefit.
Without needing to query the database, valuable data insights can get into the hands of users faster. This is especially the case for Edge applications. Applications make it easier for non-technical teams to understand what’s happening.
Edge analytics applications streamline the decision-making process: changes can be implemented based on present (not historical) conditions.
They also make operations teams less reliant on technical teams for data pulls, advanced analyses, or after-the-fact reporting. This can streamline the decision-making process so that changes can be implemented based on present (not historical) conditions.
Yes, Edge devices can be expensive. However, by not paying for ever-increasing amounts of data storage and management, Edge solutions can be cost-effective in the long run. Companies can save on bandwidth needs by analyzing more data at the source instead of transmitting everything to a centralized data center.
Beyond these tangible cost savings, enabling teams to react faster to changing performance can also save money. For example, preventing unnecessary scrap or defects can help yield an almost immediate ROI when implemented well.
Generally speaking, the further data has to travel from the point it is collected, the more vulnerable it is to hackers (all the more reason to partner with a secure, certified provider instead of using in-house data solutions!). Since Edge data exists close to the source of data generation, data doesn’t need to travel as far and poses fewer security risks.
Braincube offers Edge applications that streamline data analysis for production teams. For example, Braincube’s Counter App gives teams the ability to live monitor production using no-coding-required data visualization dashboards. Teams can stay on track with live access to production rates and defect rates without having advanced technical skills.
These applications can help level-up employees’ data literacy. With greater access to and understanding of data, employees may feel more invested in their work.
Tools like Braincube’s Edge applications make it easy to compare aspects of specific manufacturing processes as well as different plants. As a result, teams can identify—and scale—continuous improvement opportunities across facilities.
Real-time alerts automate manual processes by notifying teams about potential problems before they arise. This means teams can minimize potential unplanned downtime and, ultimately, save resources.
One of the biggest decisions that manufacturers have to make in regards to Edge solutions relates to data management. A key value of Edge analytics may also be a deficit: if Edge data isn’t stored in another system, the data disappears.
The good news is that most Edge analytics platforms give companies a choice about what to do with Edge data after it is collected; data doesn’t necessarily have to disappear. Companies must determine which data is the most valuable to them and if it’s necessary to store any of it.
Most Edge analytics platforms give companies a choice about what to do with Edge data after it is collected; data doesn’t necessarily have to disappear.
Is data only used in live time, meaning the majority of data isn’t stored? Do you want to only store key pieces of analyzed data, such as hourly averages, machine performance, or dynamic OEE scores? Or do you still want to send all your data to the Cloud even after using it in live time? Depending on what you chose, it’s likely that some—or most—of your data will be lost for future analysis.
Rarely are manufacturers choosing between a Cloud-only or Edge-only analytics strategy. The reality is that manufacturers need both Cloud and Edge capabilities.
One of the key reasons for this is that Edge technologies have their limitations when it comes to diving deeper into why something is happening. Yet as companies continue to make a case for 4.0, it’s better to start using data than just pay to store it.
The reality is that manufacturers need both Cloud and Edge capabilities.
Edge solutions are highly valuable for live monitoring or quick improvement opportunities. However, they aren’t useful if you’re trying to perform detailed root cause analysis, improve product quality, or uncover process improvements: that’s where Cloud computing comes in.
In other words, don’t expect to fully replace your current Cloud computing technologies with Edge solutions—you’ll need them both to maximize your processes.
Since Edge data is most valuable in live time, it doesn’t necessarily make sense to utilize it for every machine at your facility. Edge analytics is most valuable in situations where speed and efficiency intersect.
Edge analytics is most valuable in situations where speed and efficiency intersect.
Some industries will utilize Edge analytics more than others. For example, if your company produces a high volume of products per minute, you’d want to know if your defect rate started spiking over the course of the last hour. If you have a longer, slower production cycle, such as steel production, there may only be a few steps in the process where Edge analytics technology is beneficial.
Before jumping in, consider the most relevant situations where Edge analytics can bring value to your organization and processes.
It’s easy to confuse Edge analytics and Edge computing. In many instances, the two terms are used interchangeably. Still, there are distinct and important differences between these two types of Edge technologies.
Edge computing involves the physical aspect of collecting and processing data close to where data is generated, just like Edge Analytics. In the case of Edge computing, data collecting and processing may take place at the device level or the gateway level.
By comparison, Edge analytics refers to the collection, analysis, and event streaming of real-time IoT data. Edge analytics utilizes the same data captured by IoT devices, but it also incorporates historical data collected from a given device. Edge analytics is similar to data intelligence, in that data is transformed in some meaningful way: for example, an automatic report, an alert, or a dashboard.
Generally speaking, Edge analytics is a leveled-up version of Edge computing. By analyzing live data alongside past data, teams gain actionable insights from Edge analytics outputs that they may not get with a typical Edge computing solution.
If you’re not already using Edge, it’s time to raise your expectations about what Edge data can do for your teams and organization. Edge analytics serve as an enabler for smarter, more focused production that saves time, resources, and quality mistakes.
By embracing Edge analytics, teams have information at their fingertips that improves their reaction times. In turn, these technologies can help prevent unnecessary downtime, defects, and other bottlenecks from occurring.
Edge analytics is the process of adding intelligence—such as visualization or alerting—to Edge data.
Get answers to your top Edge analytics questions in this introductory article, including how it works and the value it brings your organization.
Edge solutions enable your teams to process data off your line and immediately put it to use. This low latency data helps your shop floor better adapt its strategies on the fly. Applications, designed for Edge data, will give your teams real-time snapshots of what is happening.
Decathlon produces high-quality, sustainable and cost-effective retail athletic products. In order to accelerate their digitalization efforts, Decathlon sought an Edge solution to provide insight into daily performance through streaming data.