Other Articles

The topic of collecting data from production machines has never been more current than this year. Whether it is the energy crisis, your need to monitor energy consumption, your customers' requirements or your parent company that has made you think about it, you should carefully consider several basic questions. Answering them will help you set your digitalisation processes correctly and start gathering data effectively, with tangible results. The first question is: Why do you want to capture data in the first place?

Collecting data from machines and production lines, be it production or technological data, is this year's most popular topic with our customers. The time has come. The Covid-19 pandemic has persuaded even the more conservative ones. It is now crystal clear that not only will plant digitalisation soon give companies that have begun this journey a competitive edge, but it will become a must.

What has actually made you want to collect data?

We are meeting companies that are planning to get started with digitalisation, modernise their operation and start capturing data because they are being pushed by their parent companies or management.

At the beginning, we always ask these companies why they need data in the first place. "When you've collected data and stored it in a database but then don't leverage it, how will this benefit you? That's why we first ask our customers what problems they have in their plants and then think about which of them can be solved by implementing data acquisition and analysis," said Michal Tůma, CSO at FOXON. "The thing is, only reliable data can be turned into information that will help you make better decisions," said Tůma.

You will probably start by implementing energy consumption or downtime monitoring

More and more companies are starting to realise that it is lack of detailed production data that hinders them from making their processes more effective and efficient. What are the needs that usually translate into the first steps on the data acquisition journey?

Some customers need to better monitor OEE and quickly detect the causes of downtime and scrap material. There are also companies that place more importance on energy consumption monitoring, as a result of rising prices of electricity, natural gas and compressed air.

And then you can find companies that are under pressure from their suppliers and customers to provide product birth certificates. By tracking and storing technological and production data, manufacturers can see both real-time and retrospectively how a particular product is / was being made (e.g. conditions on the shop floor, the materials used, etc.).

5 questions to ponder before you get started with data acquisition

Does anything from the section above sound familiar? Can you identify other problems, for example in your production, quality or maintenance? Discuss the following questions with your colleagues from IT and management. You will begin to get a real idea of digitalisation and data collection in your plant. While it is natural to start small, do not be afraid to think big from the very start, even beyond the needs of your own department.

Alternatively, you can discuss these questions with an independent consultancy. An outsider's perspective is invaluable. We will share with you our experience with our projects implemented at other plants. In addition, for us as an external consultancy, it is easier to get your employees' answers to questions about their problems at work. This is how we launch FIOT digital transformation projects.

Question 1: Are you able to read data from control systems?

You should put your first questions to your PLC programmers, maintenance engineers and IT specialists. The questions may include:

• Are your production lines network-connected?
• Do you have access to data in your control systems?
• Is access to data in your PLCs password-protected or forbidden by default by the machine supplier?
• Can you understand the data that your PLCs provide?
• Do you know how to transfer data from your PLCs to a higher-level system where you can read it, store it in a database or even display it?

Every company should have an employee solely responsible for digitalisation and machine connectivity, who acts as a mediator and tries to accommodate the needs of maintenance, production and IT. This person is responsible for having machines networked and data standardisation and cybersecurity. Their job is also to find a way to retrieve data that is hard to access. If you, for example, have an atypical control system in your production, if the data required is not available in it, or if your machine supplier has not provided you with access to this data, you need to find a way around it. The solution does not necessarily have to be complicated. Sometimes it is as easy as deploying wireless sensors.

Question 2: What data do you need?

This question is closely related to some of the ideas in the introduction of this article.

• What information do you need to be able to solve your problem?
• Do you need production data? OEE data (e.g. the number of pieces produced, OK/NOK, downtime and cycle time)?
• Or are you interested in technological data (e.g. temperatures, pressure, vibrations and the number of revolutions)?
• Does your maintenance lack information about the quality of your PROFIBUS and PROFINET networks?
• Do you know that you can learn much more?

You can also gain some fresh insights by combining several different quantities. For example, by putting together data about a temperature rise in a machine and its vibrations, you can predict an imminent failure and prevent it before it stops your production line.

The right consultancy will also help you make connections and see the full picture. You may not realise that the quality of your production may be dependent on temperature changes in your plant throughout the year. Or that materials from different suppliers behave a little bit differently, so depending on the material used you may need to modify the production process.

Let us meet and discuss this. We may even identify some other needs of yours.

Question 3: How many machines should you collect data from?

At the beginning you will probably decide to collect data from one machine or production line. This is good. A pilot project like this will give you some idea of what to expect and the benefits.

But remember that in the future the number of machines that you need to collect data from will grow, as will the number of people who need that data. For that reason, the technological solution that you choose should be universal and scalable enough; it should be suitable for the entire company. The last thing that you want is to have to replace your functional system with one that is suitable for more applications.

You may also want to consider whether you can achieve a good ROI with regard to a particular machine. If you are going to modernise part of your production line next year, deploying data acquisition makes no sense. In addition, you do not have to collect data from all of your machines. Focus on key elements of your production and areas that need to be covered.

Question 4: What do you want to do with your data next?

Storing your data in a database is hardly enough. You have to visualise and analyse it. Turn it into information that will be useful for several departments in your company:

• Maintenance staff will use it to fix faults faster and better plan preventive maintenance.
• The production department will learn soon enough if production starts going down.
• Quality will identify a bad-quality series much faster, not at the end of the shift.
• The management across the company will have one place to go to for a single version of the truth and will generate reports faster.

The goal is not to gather more data but to achieve positive outcomes – more efficient production, happier staff or more customers.

Question 5: What is your budget?

How much does data acquisition cost? To be honest, not very little. While collecting data from one machine may start at tens of thousands of crowns, capturing data from more machines will likely cost hundreds of thousands. And collecting data from the entire plant will cost you millions. The price will depend on what control systems you have, what data you want to capture and how you intend to use it.

Above all, you should keep in mind that digitalisation is not one project with a limited budget. In the same way that a factory continuously invests in machinery, recruitment, training and research, it should allocate a certain percentage of its annual turnover or profit for plant digitalisation.

You can start with a pilot project with a fixed budget (How else would you evaluate it then?), but expect your digitalisation project to grow.

Is your company financially ready for this? Discuss this topic with your management and show them that digitalisation is no longer a luxury but a must. Or approach someone and ask them to explain what to expect. Otherwise, you might soon get crushed by your competitors.