We’ll look at a real case study of a company from the automotive sector operating in Slovakia, which decided to test the potential of digital transformation even on older equipment. And the result? In the first few weeks alone, productivity increased by 20%—without purchasing any new technology, solely thanks to the availability of accurate data and increased transparency. How exactly did it work, what did the implementation involve, and what results were achieved? You’ll find out in the following paragraphs.
What were the biggest challenges the company faced before implementing OEE?
The company had already implemented an OEE system on part of its production lines, provided directly by the machine manufacturer. However, problems arose on other equipment where no automated data collection existed. Production in these areas relied entirely on paper forms, which operators filled out during or after their shifts. This data was then transcribed into Excel, reviewed manually, and only afterward entered into SAP. The entire process was time-consuming, prone to errors, and did not allow for effective analysis.
In addition to production data, information about the orders themselves—how much to produce, what product type, and the required cycle time—was also handed over manually. These details had to be printed, physically brought to the workstation, and pinned to a board so operators could see them. And if anything changed, the whole process had to be repeated from scratch.
An even greater burden was the manual recording of downtimes. Every outage had to be documented and assigned to one of thirty predefined reasons. In practice, usually only the most significant downtimes were logged—and even those were mostly estimated. The most frequently used reasons were general categories such as “cleaning” or “repair,” which failed to provide a real picture of the actual causes of production stops. Management lacked reliable data, had no solid basis for decision-making, and certainly no tool for immediately evaluating the effectiveness of corrective measures.
Why did the client decide to proceed with implementation, and what were the main goals of deploying OEE?
The company aimed to find out whether digital transformation could be carried out even on older production equipment with a Siemens Simatic S7 control system. For this reason, they decided to carry out a pilot PoC (Proof of Concept) project to test how the OEE system would operate in the real conditions of their production environment.
The main goal was to demonstrate that even on equipment that originally did not support automated data collection, it is possible to build a functional and flexible system for measuring production performance.
It was also important to determine the time and financial demands of scaling the solution across the entire production sector. At the same time, the PoC project was intended to deliver the first hard data on which a realistic return on investment could be calculated.
What was the OEE implementation process like in practice?
The first step was to design how to connect the equipment to the (Ethernet) network so that data could be read from it. Connecting older machines is often a major challenge, and in this case, one of the key project goals was to verify whether such integration was even possible.
Next came the definition of the data that the system would need for OEE purposes. This included operational, production, and failure states of the machine. The partner company, Galaxy Automation, analyzed the PLC (Programmable Logic Controller) and identified which of these data points could be reliably and safely collected from the equipment.
We then jointly defined, for each component of OEE (availability, performance, quality), which specific data would be used for evaluation and which would serve as supplementary information to improve oversight. Based on these defined data points, an application—the software part of the solution—was developed to collect data over the network, visualize it, and store historical records.
After implementation, IO tests were conducted to verify the correctness of the connections between individual data points and the overall functionality of the system. Once the tests were successfully completed, the system entered trial operation, during which any discrepancies were resolved. Only then did full-scale operation begin.
The result was a comprehensive system that collected and evaluated nearly 50 different parameters from the machine in real time, providing an exceptionally detailed view of every aspect of its performance.
Availability Measurement
In terms of availability, we monitored all machine states during which the equipment was not producing—whether due to malfunctions, safety conditions such as open doors or emergency stops, missing material, or a full output buffer. These downtimes were visualized in real time for the current shift and also displayed in shift reports. To provide better insight into long-term trends, summary overviews for 7- or 30-day periods were also available.
All downtimes lasting less than one minute were automatically categorized as short downtimes, without requiring any additional manual classification. This sorting was handled at the software level, allowing the customer to easily adjust the time threshold (e.g., from one minute to another value) without having to modify the PLC code.
In cases where a downtime could not be automatically identified, the operator was prompted to provide a reason through a pop-up window. If no reason was entered, the downtime was labeled as “unclassified,” which was clearly visible in the graphs and reports for the relevant shift. A key benefit was that, for the first time, the recorded data could be mapped down to the level of individual operators—creating pressure to improve data quality.
During the pilot project, we also analyzed how to fully automate the downtime classification process. Although this extended beyond the scope of the PoC itself, it was a crucial discovery that classification automation is not only possible but realistically achievable in the next phase of the project. This would completely relieve operators from manual input and significantly increase both the precision and reliability of the collected data.
Performance Measurement
In terms of performance, it was essential to precisely identify when a product was considered finished to avoid data distortion. We achieved this by monitoring the piece count directly at the PLC level using hourly and shift counters, which were automatically reset by the software at the start of each hour or shift. This allowed not only reading data from the machine but also actively writing data to it.
Additionally, since the machine allowed manual speed reduction, we monitored this parameter as well. The duration of slowdowns—like all other collected or calculated data—was recorded in a database and reflected in reports.
Quality Measurement
As the equipment did not include automated output control, quality continued to be recorded manually. However, in the OEE application, we created a simple interface that allowed operators to conveniently enter the number of defective pieces along with the reason for the defect. While the recording was manual, it was immediate, digital, and significantly faster than the previous paper-based method.
Finally, we replaced colorful and cluttered screens with a visualization based on the ISA-101 High Performance HMI standard. This concept emphasizes neutral colors, high contrast, and visual highlighting only of elements that require immediate attention. The result is a clear interface that reduces cognitive load for operators and improves response time to critical conditions.
Were the goals achieved? What concrete results did the OEE implementation bring?
The pilot project successfully demonstrated all intended objectives. The client obtained real data from their own technology, which gave them a new perspective on their production operations. For the first time, they had concrete numbers in their hands, enabling real-time decision-making.
Within just the first few weeks after deploying the application, productivity increased by 20%. And this was due solely to the so-called “halo effect”—the fact that employees were aware their performance was being monitored and visible to management.
Final reports also showed that the average OEE increased from the original 56% to 70%, and the average number of units produced per shift rose by 25%.
And perhaps most importantly, the company now has the confidence that digitalization is possible even on older machines. They understand their options and have a foundation for expanding the project across the entire production sector.
Why did they choose IoT Industries?
The company needed a modern, web-based solution that would also be accessible via mobile devices, would not be limited by the number of users or connected machines, and would be flexible enough to add additional features—such as electronic work requests, automated reports, or maintenance notifications.
A major benefit for them was also the ability to integrate Power BI reports directly into the application, along with our team’s experience with virtual server infrastructure and enterprise environments, modern authentication methods (e.g., SSO – Single Sign-On), and especially the integration of SAP with manufacturing technology.
Just as important, however, was finding a partner that not only understands IT but also the reality of production—industrial protocols, operational logic, and the challenges of connecting legacy equipment. Thanks to this combination of skills, we were able to deliver a tailored solution with minimal input required from the client.
The hardware side of the PoC project was handled by our partner Galaxy Automation, which ensured the connection to the Siemens Simatic PLC system via the Profibus industrial protocol and converted it to the modern and secure OPC UA standard. At IoT Industries, we then designed and implemented the entire OEE system, including the application layer built on the Ignition platform.
And what did the customer say?
The customer appreciated that we were able to independently analyze their needs, design the solution, and put it into operation without burdening them with unnecessary steps. All they had to do was define their main expectations. We took care of the rest—practically turnkey.
If you’re also considering how to make your production more efficient, obtain accurate data, and finally make decisions that have a real impact—get in touch with us. Together, we’ll find a solution that works for your production too.