How is it possible that ChatGPT has already created PowerPoints and Pivot tables for us, while we still struggle with questions about production:

• How much material and energy did Order X consume?

• How many downtimes have my machines experienced and what are the top reasons?

• When did I exceed critical thresholds?

• Are my process parameters consistent across orders?

• Am I producing better or worse than I was a month ago?

The answer lies in the lack of availability of relevant machine data throughout the company. But what prevents us from making the vision of Industry 4.0 a reality? What challenges do we face in digitalization?

This article is for you if…

Are you busy every day with Industry 4.0, connectivity/machine data, IIoT, and standards like OPC UA? Or are these topics relevant for your company?

With our experience from over 100 digitized machines from heterogeneous equipment fleets of different manufacturers and years (1980 to today), we would like to share our knowledge from the last five years in this blog series. In a total of 8 articles, we will take you into the world of machine data and connectivity, showcasing the most significant use cases we have seen and realized with customers.

For better clarity, you will find a list of all blog posts at the beginning and end of each article:

Overview Blog Series Connectivity & Machine Data:

1. OPC UA: Blessing or Curse for Industry 4.0?

2. Digitalization Dilemma: Working for the Data or Working with the Data

3. From Euromap, Data Blocks, and Harmonized Data

4. Data Harmonization, or: What is my Throughput?

5. How Edge Devices Do Not Become a Security Gap

6. No More Closed Systems

7. Ready for All Challenges in Production with ENLYZE and Grafana

8. The Key to AI in Production

Lesson #1: Plant Connectivity is Not Yet Solved

The success of Industry 4.0 crucially depends on scalable capturing of essential machine data. This differs fundamentally from traditional machine data collection. While conventional MDE modules for ME systems store data such as runtimes, quantities, speeds, downtimes, and malfunctions, Industry 4.0 requires capturing all process-critical parameters. These should be:

• in large quantities

• with high frequency

• structured and standardized

collected and stored. The focus here is on the so-called Industrial Internet of Things (IIoT) platforms. Their primary task is to make the data easily accessible for other systems, thereby creating added value for the entire company. Beyond integration into traditional ME and ERP systems, data from an IIoT platform can be integrated into other systems without substantial IT effort.

Through this networking, new use cases in the field of Big Data, anomaly detection, and Artificial Intelligence (AI) can be implemented. Moreover, traditional reporting and calculation functions of existing business software can be enhanced or replaced by modern and flexible tools like PowerBI.

The marketing promise of IIoT platforms is not only in particularly easy data provision but also in a particularly easy connection of plants. While connecting legacy systems like historians is often a lengthy project requiring many IT resources, the effort for IIoT platforms is reduced to a 1-click integration.

The central pivot of this promise seems to be OPC UA (Open Platform Communications Unified Architecture), which is often celebrated as the supposed key to unifying machine communication and connection. The goal of OPC UA is to serve as a uniform communication interface to the central plant controls and the plant peripherals.

"If the plant has OPC, then it’s not a problem."

If you are dealing with the topic of plant digitalization and IIoT platforms, you have probably heard this phrase several times. In this article, we take a critical look at the protocol and discuss the actual performance and application of OPC UA.

🧩 Not an Integral Component, but a Subsequent Add-On 

Although OPC UA is often touted as an integral part of machines and controls, it frequently turns out to be a subsequent add-on that machine manufacturers use as an additional source of income. 

We have been here before: Our experience has shown that many machines from the 1990s (long before "Industry 4.0") were originally equipped with the then OPC DA standard, and this completely free of charge.

Due to the add-on nature, OPC UA is often treated "stepmotherly". Siemens aligns itself with the OPC Foundation and provides the technology, but the implementation rests with the OEM. A lack of knowledge or relevance of a good implementation often leads to poor performance of OPC UA servers, which in some cases crash or even overload the plant control.

Often, not all variables available on the HMI can be accessed via OPC UA. Support from machine manufacturers is also limited since it is not directly related to the operation of the plant. In all these aspects, it becomes apparent that the machine manufacturer feels differently responsible for the OPC UA server than for the plant itself.

🧶 The Complexity of OPC UA: Solution or Source of Problems? 

OPC UA was supposed to be the solution for the proliferation of machine protocols, but the complexity of the system often leads to new problems. This is evident, for example, in the challenging translation from OPC DA to OPC UA. 

In reality, these are two closely related machine protocols and a theoretically straightforward implementation, but in reality, it is so complex that manufacturers often do not offer an upgrade from DA to UA.

The complexity also leads to significant performance losses. Even the most modern controls, like the Siemens S7 series (S7-1500 and 1200) or controls from B&R, reach their limits when reading a larger number of variables via OPC UA.

Many control manufacturers therefore rely on their own internal machine protocols, which perform better in terms of efficiency and performance. We at ENLYZE also prefer direct access through these protocols to avoid the problems stemming from the complexity of OPC UA while ensuring the same variables that an HMI receives.

🗣️ OPC UA is a Communication Standard, Not an Industry Standard

The presence of OPC UA servers unfortunately does not mean that they are compatible with each other. The differences range from the naming schema used to the supported encryption and authentication methods. This often results in problems and lengthy troubleshooting.

How a unified naming schema can still be created from differently implemented OPC UA servers will be explained in the third part of this series.

🔎 Critical Examination of OPC UA and Alternatives

Theoretically, OPC UA should be the solution to today's variety of machine protocols. However, in practice, it becomes apparent that the complexity of OPC UA often creates new problems, and alternative connections may be better suited.

Before deciding on OPC UA, it is essential to carefully check what is actually offered, especially concerning performance, number of variables, costs, and support. 

If OPC UA does not deliver the desired results, there are alternative machine protocols that are often simpler and offer fewer pitfalls.

If OPC UA is not the Holy Grail on the path to Industry 4.0, how can the entire machine fleet be connected? What else needs to be considered when building a comprehensive IIoT platform?

We hope this blog article has provided you with a new perspective on the challenges of plant connectivity. In the following articles, we will cover more lessons around the theme of digitalization in industry and IIoT platforms and shed light on technical challenges.

We would be delighted if you join us on this journey. In the meantime, if you have any questions, feedback, or a contrary opinion, please feel free to reach us at hello@enlyze.com. We look forward to hearing from you.