Illuminating Dark Data, Mastering Retrofit: Enhancing Plant Efficiency and Safety

Hey, have you ever heard of Dark Data? It refers to data that exists but can’t necessarily be used – often because it isn’t connected or integrated. But how can such integration be approached? And is what’s being marketed to us as a solution really something new, or is it just old concepts with new names? That’s exactly what we’ll explore today as we dive into concepts like Unified Namespace.

What does Unified Namespace really mean for digital transformation? Is it genuinely groundbreaking, or just a marketing hype? Together with Wolfgang Laufmann, Business Developer Smart Factory at WAGO, and Hans Otto Weinhold, Principal Solution Architect at AVEVA, we take a deep dive into this topic. It’s all about Unified Namespace: Can a centralized real-time data architecture help here? What is the MQTT broker all about and why is this protocol becoming increasingly important? Also: What do you need to bear in mind when implementing it?

Why are WAGO and AVEVA now working hand in hand, and how can you perhaps become a part of it yourself? This episode is also exciting for anyone who has heard the name OSIsoft PI System before. If you’re hearing the term Unified Namespace for the first time today, make sure to bookmark Episode 122! In that episode, we talk with Frank Thelen and the founder of United Manufacturing Hub, Alexander Krüger, about this very topic – it’s definitely worth it!

As always, you can find all information about the implementation and today’s use cases at www.iotusecase.com and in the show notes. And with that, let’s go! Enjoy the episode!

Hi Wolfgang and hello Hans Otto, welcome to the IoT Use Case Podcast! I’m thrilled to have you both here today. Wolfgang, let me start with a question for you: How are you doing? Where are you right now?

Wolfgang

I’m fine, thanks for asking. I’m from Hagen – some might remember the 80s when Hagen was known as the “Become-a-Popstar City.” But unlike Extrabreit and Nena, I didn’t become a popstar. I’ve been with WAGO for 25 years, working in the process industry sector, and today I’m joining from my home office. My focus is on the region around the Chempark in Leverkusen as well as customers in cities like Cologne, Düsseldorf, Oberhausen, and Essen.

Great! I just googled Hagen – it’s a large city in North Rhine-Westphalia, right?

Hans Otto

Yes, exactly.

All right. Great to have you with us today! We’ll talk more about you shortly. But first, over to Hans Otto: Just a quick question – should I call you Hans or Hans Otto? Which do you prefer?

Hans Otto

Hans Otto is great.

Nice. A warm welcome to the podcast as well; it’s great to have you here. Where are you at the moment?

Hans Otto

I’m from a small village near Bad Hersfeld, with about 600 inhabitants – very rural. You might hear a dog barking or a rooster crowing in the background; that’s all part of life on the farm here. People might know Bad Hersfeld because of the big Amazon site along the A4 motorway. Today, however, I’m on the move again as I’ll be flying to a conference in Rome later. But for now, I’m also working from my home office.

Interesting! Have you ever been to the Amazon logistics center there? I’d love to hear about it because I’ve only ever seen it from the motorway, as you mentioned. Have you had any contact with it?

Hans Otto

No, actually, I haven’t. But I’m currently working on a project with a customer in Austria who produces high racks for Amazon worldwide.

It’s great to see you’re from the region. Now, let’s quickly turn back to both of you: Wolfgang, you’re with WAGO. Could you briefly explain what exactly you do there and what your business unit focuses on?

Wolfgang

I work in Business Development and am responsible for IoT. What does WAGO do? WAGO is a global manufacturer with roots in electrical connection technology, which established itself early in the field of automation, particularly with the WAGO I/O System. This business unit has been continuously expanded. In Business Development, I act as the interface that captures customer requirements and market trends, passes them on internally, and ensures that the right products are developed and provided.

It’s particularly exciting that you’re strongly focused on IoT products and solutions. How did the connection to Hans Otto come about? How did your companies find each other?

Wolfgang

We are constantly developing our product portfolio and often act as an OT-IT gateway. In our projects, we frequently connect to existing control systems or develop solutions for ERP systems. In doing so, we like to collaborate with technology partners. We’ve known Hans Otto for several years now. Through collaboration with mutual customers and projects, a technology partnership has evolved, which I’m happy to integrate into our product solutions. Both sides benefit from it.

Hans Otto, perhaps you could tell us a bit about AVEVA and how your portfolio complements this partnership so we can better understand it.

Hans Otto

AVEVA is a software provider specializing in OT, Operations Technology, and ET, Engineering Technology. We are well-known for covering a wide range of applications with our software. Founded in 1967 as a spin-off from the University of Cambridge, AVEVA initially focused on Engineering Technology. For instance, we developed one of the first true 3D programs for CAD and CAM, which are used for planning cruise ships, large buildings, and their corresponding engineering.
In 2018, Schneider Electric transferred its entire software division for OT and Data Management to AVEVA. I joined AVEVA in 2021 after the company acquired OSIsoft from the USA. I had been with OSIsoft since 2004, for over 20 years. We are particularly well known for the PI System, a process data management solution that is widely used in industries such as chemicals, pharmaceuticals, oil and gas, and power generation and distribution. Today, AVEVA has about 6,500 employees and is a wholly-owned subsidiary of Schneider Electric.
Why do WAGO and AVEVA work so well together? On the one hand, one might assume there’s some competition between Schneider Electric and WAGO. However, AVEVA deliberately positions itself as an open and independent platform accessible to many market participants. Our data infrastructures, such as the PI System or the SaaS CONNECT platform, are designed to enable access to as many third parties as possible.
The first interactions between AVEVA and WAGO began in 2017 through a mutual customer. This customer wanted to perform a retrofit for their chemical park, using WAGO controllers while enabling direct data connectivity to the PI System without setting up a separate data acquisition system. The goal was to integrate data directly from the field level into the PI System via edge computing. This collaboration led to the development of a solution that met these exact requirements.

That’s impressive. Many of our IoT-enthusiastic listeners are always keen to hear a concrete use case. I understand it’s often difficult to get customer approvals, but perhaps you can still share a project that illustrates the dynamics you operate in and how you tackle challenges together. Is there a suitable example you can share?

Wolfgang

Digitalization and IoT cover a wide range of applications, including energy data management, monitoring, infrastructure automation, predictive maintenance, and more. For today, we’ve chosen the example of safe tank loading, which integrates all these aspects.
Safe tank loading involves two key components: first, capturing fill levels and the quantities to be loaded directly at the facility, and second, integrating this data into an ERP or SAP system to manage the entire workflow.

Quick clarification: When you talk about tank loading, do you mean silos? Or tanks in trucks? Or something else?

Wolfgang

It’s versatile. Tank loading can be a classic silo with a truck underneath or a large loading station for railway wagons. The term encompasses all these scenarios.
In addition to monitoring fill levels and recording the quantities loaded, it is crucial to ensure proper allocation—meaning the correct substance is loaded into the right tank truck. Mistakes here could lead to significant incidents. Furthermore, the billing workflow is integrated into the SAP system.
Another important element is the use of intelligent camera systems that can detect anomalies during the loading process. For example, they can spot issues such as jerking valves or leaks. These processes are often automated, especially in explosion-prone areas where no personnel are present to notice such anomalies. Cameras can detect and evaluate anomalies directly on the equipment level and, if necessary, halt the loading process to prevent incidents.
In short, it’s about protecting the environment and people while ensuring smooth operation of the facility.

Great! That’s truly a case we see frequently in our community. If you’re listening and thinking, “Tank loading isn’t my area,” you can still view this use case as an example, as it translates well to other applications. There are many similar projects in our network, and it’s exciting that we’ve chosen this one today.
In the context of tank loading, it’s about both condition monitoring with links to maintenance and how data acquisition is implemented technically, especially in retrofit scenarios. Perhaps we could delve a bit deeper into the business case. You’ve already touched on this: customers make significant investments here. What savings are possible, and why is it worthwhile to invest in IoT technologies?
What are customers—exaggerating a bit—currently losing in terms of time and money? And what are the key challenges in the business case?

Hans Otto

When implementing such systems, it’s essential to assess whether the investment is worthwhile. Among our customers, we often encounter a variety of different systems that don’t communicate with each other. We call this the “flying spaghetti monster”: there are numerous data sources, including SCADA systems, Excel files, or controllers where data is collected but not directly integrated into the SCADA system, as it’s often not necessary for process control.
As a result, maintenance teams rarely have automated access to the data they need and only act when urgent action is required. Our approach—using tools like the PI System or retrofit solutions with controllers—is to make this data accessible to maintenance personnel and other user groups within the company.
Looking at the safe tank loading use case, we often see customers relying on manual processes, such as paperwork: data is recorded manually, which can lead to errors like typos. What’s missing is a reliable data foundation that can be trusted and whose data can be further used in ERP systems.
When automating this process, several elements come into play. For instance, license plate recognition of the truck that arrives for tank loading. Additionally, all information is verified: Is the tank loading proceeding as planned? Are all conditions met? What material is being loaded or unloaded, and in what quantity? The PI System consolidates and contextualizes data from various sources.
On the SCADA or control system level of a tank facility, for example, you might see which materials are stored in tanks, as well as temperature or pressure readings. However, the context—what is being loaded or unloaded—is often missing. This context can be created through retrofitting, by tapping into additional data sources and including them in the analysis.
With automated documentation of tank loading processes, the business case often becomes clear very quickly. You start to identify relationships that were previously unrecognizable due to the lack of accessible data. The key is integrating various data sources and data pots, unlocking significant optimization potential.

Absolutely! I also checked our community database to see how other customers have tackled similar use cases. Fleet management, for example, is a significant area. Many customers lose substantial sums due to underutilized truck fleets or unnecessary trips. We’ve seen a project where up to 2,000 trips per year could be saved through better planning.
Inventory management and digital reordering are also critical topics. High costs often arise from overstocking or inefficient reordering, which better planning could avoid. This represents a valid business case with substantial potential.
Earlier you mentioned a project in a chemical park where the aim was to avoid having to collect any additional data. Could you tell us more about that? What are common challenges or mistakes customers make in such projects? Where are the technical obstacles?

Wolfgang

One key factor—and this was briefly touched on earlier—is the human element. While we are all human, the goal should be to automate this factor as much as possible to achieve greater efficiency and safety.
An example outside of tank loading would be applications where truck loading is documented. Such processes are often carried out manually with photos in order to avoid damage or subsequent damage. Automated solutions can be much more effective in such scenarios.
A major challenge in these projects is often the lack of staff on-site, with employees covering broad areas of responsibility. When processes are predominantly manual, errors inevitably occur. These can lead not only to disruptions but also to complete plant shutdowns. This is a critical area for optimization.

Was that the same technical challenge your chemical customer faced? Specifically regarding data acquisition—you mentioned earlier that no additional data should be collected. Does this align with that, or was it a different challenge?

Wolfgang

That definitely falls into the same category as what Hans Otto described earlier: ensuring the documentation of correct, reliable data directly from the field. Often, inaccurate or incomplete data is available, making optimization efforts difficult or even impossible. Accurate real-world data is the foundation, and it must be captured automatically. Without automated data acquisition and allocation, no sustainable optimization process can be developed.

Absolutely. I can imagine that interface management is also a significant challenge—after all, the data needs to be integrated into ERP or other IT systems, as you both mentioned. Hans Otto, that’s one of your areas of expertise, isn’t it? Is interface management a huge challenge?

Hans Otto

Yes, there are many different interfaces, especially with legacy systems that often date back 20, 30, or even 40 years. On the other hand, we have modern approaches like OPC UA or MQTT, and the challenge is to bridge these two worlds.

On the AVEVA side, particularly with the PI System or the system formerly known as Wonderware, we offer many native interfaces that are out-of-the-box compatible with process control systems, SCADA systems, and similar technologies.
However, when entering a scenario where unified data capture is required—such as in facilities that previously had no data acquisition—the question arises of how to retrofit data acquisition. One option might involve additional software and industrial PCs to process the facility’s protocols. A more efficient alternative, however, is the WAGO PFC.
With the WAGO PFC, data acquisition can take place directly on the edge. The WAGO controller captures the facility’s protocols and transforms them into standard protocols like MQTT or OPC UA. These data can then be transmitted via a standard interface, running as a Docker image directly on the controller, into the PI System or our CONNECT Data Services, a SaaS solution for OT data management.

Especially in retrofit scenarios, this solution significantly simplifies data acquisition. The fewer additional components that need to be integrated, the less complexity there is in the entire process.

Wolfgang

I’d like to briefly add something here. In many projects, we often see the assumption that a standardized solution can simply be implemented. However, especially in the retrofit sector, there are rarely standardized systems. Every facility comes with its unique challenges, and this is a crucial point.
To build on what Hans Otto mentioned: On the OT level, we prepare data specifically tailored to the unique requirements of each facility. These processed data are then passed on to the IT level in an appropriate format.

Thank you very much! I’m often asked how other companies implement such use cases, what best practices exist, and where potential pitfalls lie. Why do such projects sometimes fail, and what are the key factors for making them successful?
Additionally, I have a question that’s been on my mind: Success presumably depends heavily on the type of data being collected, doesn’t it? For instance, there are classical data points like fill levels, which need to be monitored continuously to enable real-time responses. But there could also be consumption data indicating when a reorder is necessary—data that would then require integration into an ERP system.
Could you elaborate on this? Do you also see it that way, that the type of data significantly influences the solution approach?

Wolfgang

To add a perspective from the OT side: At WAGO, we make a clear distinction between the OT and IT worlds. In the OT domain, we often deal with existing systems that come with a variety of standard interfaces that need to be supported. For this, we offer flexible I/O cards with various interfaces that can be customized to fit specific needs.
Additionally, new sensor technology is frequently integrated into the system, bringing new interfaces. These interfaces need to be appropriately processed as they become more intricate.
That’s my overview from the OT perspective. I’ll now hand over to Hans Otto, who can perhaps expand on this from the IT side.

Hans Otto

One of the main concerns for our customers at the moment is the consolidation of OT and IT data. On one side, there is a strong desire to leverage data from control systems, SCADA levels, or retrofit controllers. Customers often attempt to connect these diverse data pots, and the cloud is frequently used for this purpose. OT data is captured and transmitted to the cloud.
This ties in well with the topic of challenges and pitfalls. Many customers try to develop their own data management solutions for the cloud based on various data silos. There are numerous examples where we have responded with our Connect Data Services cloud infrastructure. The goal is to integrate OT data into a traditional IT system. However, data format incompatibilities are a frequent issue. On one side, we have continuous process data such as fill levels and temperatures—classical measurement data. On the other side, there are transaction-based data models, which are best organized in relational databases. Bridging these two models presents significant challenges.

Large hyperscalers often entice IT departments with the promise of unlocking new business opportunities, leading to extensive data management implementations. IT departments are growing and data scientists are demanding high-frequency sensor data at millisecond intervals in order to generate added value.
Our approach differs fundamentally: we utilize OT data within the PI System and channel it into our CONNECT Data Services, a SaaS cloud specifically for OT data. This enables data analyses up to predictive maintenance. At the same time, we provide seamless integration options with IT systems on the IT side, such as Databricks, Snowflake, and similar solutions.
By bridging data integration from the field level, through control systems and data management in the OT domain, to comprehensive data management in the IT realm, our approach becomes particularly attractive to customers.

So, your statement is that certain customer segments are better off not building their own data management platform but rather relying on partnerships. You’re experts in this field, and a standard cloud solution from hyperscalers often isn’t enough. Different data types and their optimal storage need to be considered so they can be accessed for various use cases. That’s your point, right?

Hans Otto

Exactly. This is also the focus of my work in what we call Lighthouse Projects, where I work as a Solutions Architect at AVEVA. These projects began when we were still OSIsoft and wanted to launch our OSIsoft Cloud Services. The goal is to evaluate, together with customers, over innovation projects lasting 8–12 weeks or sometimes longer, how use cases can be implemented, the underlying business case, and whether the technology works as planned.
WAGO is a valued partner in these projects, especially in the area of energy efficiency. With EDM, Energy Data Management, WAGO offers a solution for capturing energy data, which can be directly integrated into our systems and made available to us.

This is the perfect transition! To wrap up, I’d like to ask you a few more questions about how exactly you implement these solutions. You’ve already mentioned some key points, like your SaaS CONNECT Data Services. If I’m a listener now and say, “I want to implement this use case with you and discuss it further,” how would that work in practice?
Can you explain what a complete solution would look like? Of course, customers don’t have to use every component, but it would be interesting to understand what you offer—from data acquistion to integration. Let’s start with data acquisition. Wolfgang, I assume this is where WAGO comes into play with your hardware, combined with the IT infrastructure to transmit data via interfaces. How does that work exactly?

Wolfgang

Exactly, it’s actually quite simple and structured. The first step is to capture the available data, especially in brownfield facilities. We start by creating a list of the existing data and identifying which interfaces are needed. From there, we conduct consulting sessions where we work with the customer to determine the appropriate interface cards and solutions.
Often, we can rely on ready-made solution packages that we offer at the PFC or Edge Computing level, including features like an integrated data logger. This allows us to quickly move into detailed discussions with the customer. What’s special about our consulting sessions is the close collaboration: we bring both our OT expertise and the experiences and solutions of partners like AVEVA to the table. With pre-packaged solutions like EDM, we can often start right away. EDM, for instance, is a ready-to-use application in a box that can be easily retrofitted to existing systems.
In further discussions with the customer, we collaboratively develop a tailored solution that meets their specific requirements. This process creates an active dialogue.

That’s incredibly exciting and really highlights the strength of your role and expertise at WAGO. You’ve implemented so many different projects that you can provide recommendations for any use case—from the right hardware to Edge preparation and integration with partners like AVEVA.
For all listeners: If you’re curious, feel free to check out WAGO’s profile on our platform, iotusecase.com. You’ll find numerous exciting projects there. Even if your use case isn’t about fill levels but something similar, it’s worth having a conversation with Wolfgang and Hans Otto. You can benefit from their experience and develop your use case together.
One follow-up question: Data acquisition takes place via your controllers and hardware, which preprocess historical data at the Edge, correct? How does the handoff to the IT world work, and what product does AVEVA bring into play here?

Hans Otto

At AVEVA, we have a broad portfolio, not only with the PI System but also with the AVEVA System Platform, which many still know as Wonderware.
When data comes from a controller or a process control system, it is stored in the PI System, which consists of several components. One of these is the Data Archive, a classic historian. However, the PI System goes far beyond a simple historian that merely collects data like a black box.
With the Asset Framework in the PI system, data can be contextualized: Where does it come from? What does it mean? In which unit was it recorded? Users can also define events, alarms, and notifications to make optimal use of the data.
In addition, with our Connect Data Services  we offer a cloud solution that enables seamless integration of the PI system or the System Platform. However, data can also be transferred directly from edge computers such as the PFC200 or PFC300 to the cloud. This edge-to-cloud integration is particularly exciting.
Often, we find that either WAGO or AVEVA is already present at a customer site—sometimes even both. However, the challenge is that departments within the customer organization often don’t communicate with each other. For example, in logistics, one customer once remarked, “Logistics is war.” Different departments build parallel infrastructures instead of collaboratively using existing solutions.
Frequently, the question arises: Why acquire a separate application for a specific area when the PI System is already in place, and the data just needs to be contextualized and passed on? These are situations I often encounter in my Lighthouse Projects.
An example includes a manufacturer of Amazon shelving systems, as well as a large mine in northern Sweden where we analyzed a conveyor belt that frequently fails in the harsh winter. The focus there was on capturing the right data and generating value from the platform.
The combination of our two companies as strong partners provides a very interesting and impactful solution in such projects.

Thank you for bringing that up again! I can only encourage everyone to connect with Hans Otto and Wolfgang. I’ll include your LinkedIn profiles in the show notes—just stop by and reach out.
We didn’t dive too deeply into the technical aspects today, but I’ll add additional links in the show notes so you can explore further. For today’s brief overview, this was already a fantastic introduction.

Perhaps one final question to wrap up: What does the future hold for the next few years? What can we expect by 2025 and beyond? Can you share any insights—be it about new technical features or your partnership?

Wolfgang

From our side, we are planning to bundle so-called Solution Packages in a more application-oriented way and offer these as ready-made sales packages. The goal is to make it as easy as possible for customers to adopt the solution.
We often hear questions like: “Is this relevant to my use case?” or “How much does it cost?” With solution packages, we can answer these questions directly by saying: “For this use case, you need hardware XYZ, and this is the price.” This enables us to kick off projects much faster and provide targeted support to customers.

Cool! So, you’re creating packages for specific use cases like fill-level monitoring or other common scenarios. That makes it easier for customers to find a solution that fits their needs. Very exciting!

Wolfgang

Exactly. This primarily relates to the field automation level and the edge computing level. With our scalable products, we’re already implementing pre-configured applications that process and pre-prepare values. Additionally, our WAGO Solution Platform offers a layer that’s especially helpful when dealing with a large number of controllers in the field—for example, for maintenance optimization.
A common scenario is updating the firmware of controllers. Through the platform, all controllers can be displayed centrally, and changes can be made with the push of a button. IT security requirements are also addressed, such as remote access to networked controllers—a huge area that’s becoming increasingly important.

You also work closely with partners in this area, right? You have a network of various partners, from Portainer to smaller and larger providers, who supply building blocks and strengthen your partnering approach.

Wolfgang

Yes, exactly! That’s actually a topic in itself and could fill an entire podcast. Partnering is a major strategic focus for us.

Wonderful, thank you! I found today’s discussion truly fascinating—both the business and technical cases. It’s clear how your comprehensive solution works, from data acquisition and transmission to the cloud level. Your partnership is truly impressive, and I’m looking forward to seeing what’s next in the coming years. Maybe we’ll do another episode next year with updates on some more exciting projects.
Thank you both for joining us today! With that, I’d like to give the final word to you.

Hans Otto

For me, this was also very exciting, as it’s my first time participating in a podcast. I really enjoyed it, and I’d gladly do this again—especially if we can delve deeper into our projects or collaboration with WAGO next time. Thank you also to the listeners for their patience and for sticking with us until the end.

Wolfgang

I can only agree with that. I also had a great time, especially with this group. A big thank you to the listeners as well. We hope we were able to bring some clarity to this complex world and perhaps inspire some deeper thoughts about these topics.

Thank you both, and take care! Wishing you a great week. Bye!

Hans Otto

You too, bye!

Wolfgang

Bye!