The Client
Inoxsys is a leader in the manufacturing sector for industrial products in Bulgaria with more than a hundred employees. The company produces quality tanks, vessels, heat exchangers, and bulk-handling systems for many industry-leading companies all over the world.
The Challenge
Though they already had an established and well regarded product range, Radomir Rashkov, the CEO of Inoxsys, wanted togain a competitive advantage by understanding the way their products perform once put into operation by their clients, so that he could target areas to further improve the product. Radomir’s idea was to introduce a system to facilitate predictive maintenance.
Based on our vast experience with industrial services, unexpected failures have almost always led to unplanned delays. On the other hand, we looked for a way to stand out from the competition and Predictive Maintenance was among the best ideas.
The problem statement was simply stated by Radomir:
Inoxsys is a purely mechanical engineering-based company and we don’t have the capability to do IT infrastructure. We had no idea how to do that properly.
The biggest challenge with predictive maintenance, however, is that it is all about gathering data, analyzing it, and making use of it. These were the factors that drove Radomir to look for an external solution provider. Radomir found that there were few options from which he could choose. However, Cosmos Thrace was recommended by a company that he trusts. After speaking with us, he commented that it was our commitment towards the challenge that was decisive in choosing us as his preferred vendor for this project.
The Solution
From day one I felt not in the role of pushing but in the role of being pulled in executing the project.
Radomir Rashkov, CEO of Inoxsys
The Client
Inoxsys is a leader in the manufacturing sector for industrial products in Bulgaria with more than a hundred employees. The company produces quality tanks, vessels, heat exchangers, and bulk-handling systems for many industry-leading companies all over the world.
The Challenge
Though they already had an established and well regarded product range, Radomir Rashkov, the CEO of Inoxsys, wanted togain a competitive advantage by understanding the way their products perform once put into operation by their clients, so that he could target areas to further improve the product. Radomir’s idea was to introduce a system to facilitate predictive maintenance.
Based on our vast experience with industrial services, unexpected failures have almost always led to unplanned delays. On the other hand, we looked for a way to stand out from the competition and Predictive Maintenance was among the best ideas.
The problem statement was simply stated by Radomir:
Inoxsys is a purely mechanical engineering-based company and we don’t have the capability to do IT infrastructure. We had no idea how to do that properly.
The biggest challenge with predictive maintenance, however, is that it is all about gathering data, analyzing it, and making use of it. These were the factors that drove Radomir to look for an external solution provider. Radomir found that there were few options from which he could choose. However, Cosmos Thrace was recommended by a company that he trusts. After speaking with us, he commented that it was our commitment towards the challenge that was decisive in choosing us as his preferred vendor for this project.
The Solution
From day one I felt not in the role of pushing but in the role of being pulled in executing the project.
Radomir Rashkov, CEO of Inoxsys
Tools Used
Power Bi
Cloudfare
TEPAKOM
Results
1.
The sensors showed an area where the machine could be improved to make it more efficient and reliable.
2.
Live data available for power consumption, loading andtemperature identifies where there is a problem or will be a problem.
3.
Project delivered on time, on budget, and meets all quality metrics.
The Solution
Other companies told us it is impossible.
Once commercials were agreed,we started the project. We had a timeline of 20 days to have a working solution in place.
The solution we developed was based on incorporating Artificial Intelligence into the machines to enable predictive maintenance patterns that would alert operators when a machine is about to fail. To facilitate this, we were able to quickly and easily install Teracom sensors on the machines to monitor and extract data from them. Then to transport the data from the sensors transportation and into storage we used Cloudflare. The main reasons for this choice were, their serverless infrastructure ensures low latency and high availability, and that they have a Data Center in Sofia not far from the company’s plant. This ensured short response times and timely reliable data. We then used PowerBI for analytics. To ensure that we have all data securely backed up we have a separate PostgreSQL database.
An architectural diagram of the infrascrutrue:
Each sensor collects and measures specific data such as temperature, electricity consumption, etc. Then Power BI is used to visualize the data. We chose Power BI for its powerful visualization abilities, its accesses and authority features, along with its easy-to-use and understand interface.
So, once the whole system was installed and configured, we quickly started collecting, transferring, and analyzing essential data from the machines. We managed to complete the entire project as planned in 20 days, which enabled Radomir to show the working concept to the vendors as planned. According to him:
The project proved a concept that we can deliver benefits from monitoring equipment. It hit the goal we had in mind and there is a lot of demand for it in the manufacturing sector where assembly lines are incorporated.
The Result
This was a successful project, for Inoxsys. They are now able to use the Cosmos Thrace AI Solution to track live data from the machines and clearly see when a machine is being overloaded and about to fail. Along with data on how much electricity it is consuming, and whether it is working optimally, in close to real-time. Additionally, shortly after the fulfillment of the project, Radomir got a very useful insight regarding one of the machines we were tracking:
We saw that based on the data analysis the motor gearbox of a machine was working inefficiently – the machine was designed wrongly but we figured it out once the sensor was installed and data was received.
Screenshots with live data from Power BI Graphs:
Conclusion
Overall, this was a fascinating use case. The short time frame was a challenge that few other solution providers wanted to undertake. However, Cosmos Thrace was able to understand the problem, design the solution, and implement within the allowed timeframe. We also met our primary objective of delighting our customer. We were very glad to hear that based on our solution, Inoxsys plans are now focussed on:
The transition to a more high-tech company than we are at the moment, which can give us an edge over the competition.
Do you have an “impossible” challenge for us? Give us a call and allow us to achieve the impossible for you!