Global Business and Economic Growth

Getting the most from data on the edge

March 4, 2025
Artificial intelligence (AI) and edge computing increase the power of real-time data analysis

The only constant in today's industrial landscape is the rapid evolution fueled by digital transformation and artificial intelligence (AI). It is estimated that more than 90% of businesses engage in some form of digital activity. While innovation and adaptability are keys to staying ahead, the digital transformation journey for industrial organizations is by no means an easy one, particularly in water and wastewater. Control talked with Alejandro Trejo, Offer Manager for Digital Services, and James Redmond, Global Offer Manager for SCADA and Telemetry, of Schneider Electric to get further insight into the challenges and solutions for operators.

Get your subscription to Control's tri-weekly newsletter.

Q: Why is real-time data analysis and increased automation needed in industry, particularly water wastewater?

Alejandro Trejo: In today's world, there is a need to make the most of our resources, especially those related to water. To make better decisions, we must consider two relevant things. The first is the information available for us to understand what’s going on with the process and comes from a SCADA system. The second is the data that we gather must be understood as quickly as possible to make better decisions that comply with the goals and to be done with extreme accuracy.

Accuracy is provided by considering all available variables, and the need for it only increases with time. In summary, there is a need for real-time data analysis because the amount of data that we need to make the most of our resources is increasingly high. So, the accuracy of our decisions becomes more relevant as times goes by.

Q: Why does industry under-utilize data? 

AT: I think under-utilization of data is the equivalent of having the power to fly but decide to crawl from one point to another. There are a couple of reasons for under-utilization of data in our industry.

The first is PLCs do not have the required memory and power to run advanced AI applications. Many AI models require intense computational power, often requiring substantial resources to train and execute tasks in an efficient way. The second is data granularity—the level of detail available in a set of data.

AI models demand large volumes of data to ensure accuracy and high performance for tasks such as abnormal event detection and pattern recognition, data granularity and frequency play pivotal roles. Without sufficient data, these models may overlook critical anomalies, compromising effectiveness in real-world applications. So that's my comment on that, wonderful.

Q: How do AI and edge computing impact the ability to leverage data and automation not only in water, but also industry in general?

James Redmond: We see it as impacting water, energy, and oil and gas, for instance, which is the other major industries for telemetry. There's a desire to gather additional data and to turn it into information to drive more efficient operational decisions. For instance, we see in oil or gas production edge computing being used to reduce lift costs, determining preventative maintenance routines, optimizing the deployment of personnel, or limiting travel time to these remote sites. These are recurring concerns. 

What AI provides is the ability to automate this analysis and to drive better decisions. You can project when a technician is required to go out to a site to adjust the valve or to move the counterweights rather than someone driving around on a rotation. Labor shortages in oil and gas industries are a major concern. Automated decision making allows for more efficient deployment of personnel.

AI has a massive demand for computing power, and what we see is an increase in more effective algorithms that can be then produced using that master processing power and put into smaller computing impacts. So, we can drive this information decision-making directly to the edge, and better shape the data to free up resources at the SCADA level, where this information is being processed exactly right.

Q: What role do SCADA systems play in the water industry and how is it evolving as AI comes on board?

JR: That's a very important question. Traditional SCADA systems obviously need to gather the data and apply it relatively simply. Traditionally, they are quite effective in terms of when something's above or below a certain threshold, creating an alarm, and driving an action. Now, with the greater processing power at the edge, we gather more data, and then we kind of create tension. A plant isn't generally constrained by what type of data it consumes. It can pull out everything that a PLC or edge computer can provide. The question is do you process the data at the SCADA level? We see some industry trends where there's a desire to push more of that processing into the SCADA or use it to pre-process data and then pass it to a tool further up the chain, residing perhaps in the cloud or in premise specifically designed to analyze it. 

If you look more at the telemetry aspect there are two additional challenges outside the plant. You must consider the security of the data you're transmitting and the cost and bandwidth restrictions you're under while moving the data. 

What is clear to me, though, out of all of these competing trends or needs, is that the expectations the industry has for visualization and presentation capabilities of the SCADA, are increasing. Users demand tools to create mechanisms to present the data or correction to present the products of the data as information to their users, operators, and supervisors.

Q: How can organizations be ready for all these changes?

AT: There isn’t a magic wand for us to get industry ready, but there are some actions that companies might start looking at to prepare for the technology. First, consider the power of edge systems. I know that cloud solutions are powerful and very common, but the cloud is not for every aspect of our processes. We might start with powerful, yet accessible, solutions on the edge to harness the potential data from connected products in our premises. Edge solutions are powerful and safe, and our cloud component can be an element of reinforcement for data processing, providing our industries with a higher level of insights and AI driving solutions. This hybrid approach is probably the best way to bring industries from zero to 100 in a safe and smart way.

JR: Those looking ahead to how they can best take advantage of these trends to improve their operational effectiveness should have a clearer idea of what they're trying to accomplish and what would be a meaningful improvement as they're looking to conduct an experiment or an exploration, or to create a prototype of a possible system. I think that kind of focus makes the experiment or the exploration a lot more successful. 

Q: Everyone wants to maximize value. What is the value proposition for this kind of hybrid approach?

AT: We have different solutions designed to maximize the value of data in our industries. For example, we have solutions specifically designed to gather information related to environmental conditions from a pump. The value on this type of edge device application is it is connected to our cloud service where we process the data to provide our customers with a single user interface, and you can see data in near real time. 

The value is to be able to process as fast as we can in edge devices and reinforce that data processing with AI-powered solutions. With this type of solutions, you are one step ahead of any kind of failure you might have in the facility. This type of architecture lets us process both at the edge and in the cloud, while the users receive real value from its connected products. 

JR: I think the one thing I’ll add is our ability to provide the seamless integration across a series of tool sets to provide that information right from the edge to the control room. We also have mechanisms to bring that information directly to the technicians, operators, supervisors and engineers.

Sponsored Recommendations

Make Effortless HMI and PLC Modifications from Anywhere

The tiny EZminiWiFi is a godsend for the plant maintenance engineers who need to make a minor modification to the HMI program or, for that matter, the PLC program. It's very easy...

The Benefits of Using American-Made Automation Products

Discover the benefits of American-made automation products, including stable pricing, faster delivery, and innovative features tailored to real-world applications. With superior...

50 Years of Automation Innovation and What to Expect Next

Over the past 50 years, the automation technology landscape has changed dramatically, but many of the underlying industry needs remain unchanged. To learn more about what’s changed...

Manufacturing Marvels Highlights Why EZAutomation Is a Force to Be Reckoned With

Watch EZAutomation's recent feature on the popular FOX Network series "Manufacturing Marvels" and discover what makes them a force to be reckoned with in industrial automation...