A NeuroSYS expert advises how to start analyzing production data and improve work efficiency easily.
Digitalizing Production: Key Findings
- How to start analyzing production data with limited time and financial resources?
- How to improve work efficiency and reduce downtime?
- How to implement technological optimizations with high ROI?
- What effects and improvements can be expected from the first optimizations?
Interview with Tomasz Golan, R&D Manager of NeuroSYS
In the following interview, Tomasz Golan shares his observations and examples from implementing digitalization projects in various factories. He presents practical tips for factory managers to use technology to generate savings and increase work efficiency.
NeuroSYS specializes in conducting technical audits to identify problems that reduce factory production efficiency and build modern IT solutions, including those using AI technology. Our experience and knowledge allow us to effectively support manufacturing industry leaders in optimizing processes, reducing downtime and improving operational efficiency.
NeuroSYS: What are the current trends in digitizing production, and what are the biggest challenges manufacturing leaders face?
Tomasz Golan: One of the biggest challenges is the lack of automatic data collection. Many factories still rely on manual processes, even though technologies such as IoT (Internet of Things) and various sensors are widely available. This leads to errors and inefficiencies.
Factory leaders face the challenge of deciding whether to focus on quickly launching new production lines or investing more time in analyzing and refining existing processes through data collection. Often, the time and financial pressure are very high, and ultimately, companies focus on starting production and ensuring its appropriate quality. Matters related to analytics receive a lower priority.
This is because IT, analytics, and data management are not the core competencies of manufacturing leaders. They must delegate these competencies externally, which further complicates the process.
What problems does the lack of digitization in production processes cause in factories?
These problems are complex and strongly dependent on the type of production. I can use an example from one of our projects, which shows a challenge often occurring in production plants.
We worked with a factory where operators had to manually enter data on the number of units produced and machine downtime. This information was later transcribed into Excel spreadsheets, which created the potential for errors. As a result, managers needed a full picture of the situation and could not make accurate decisions regarding process optimization.
Read also: Key Findings from Technical Audits in Factories. Interview with NeuroSYS CEO
How should factories start digitizing production? Is it necessary to begin with advanced IT solutions?
This is quite a common myth that we want to debunk by working with our clients. There is no need to open a time-consuming, highly advanced computerization process to achieve significant benefits in work efficiency.
Let’s start with small steps. First, factories can start collecting basic production data from existing PLC controllers directly from robots or their already-established vision systems. PLCs are controllers that already collect a lot of data in factories, such as the condition of machines or the number of units produced. Just capture this data and save it in the central database. This in itself can provide valuable information. However, this basic production data may also come from other equipment already operating on the production line.
Even basic automation of data collection can yield tangible benefits, thanks to the possibility of a more detailed analysis of production processes.
What benefits can such a first simple step in digitizing the production process bring?
Even basic automatic data collection can provide tangible benefits by enabling more detailed analysis of production processes.
First, we can find out where machine downtime comes from. Collecting machine health data allows us to identify patterns of downtime. For example, if we have ten identical machines and one experiencing more frequent downtime, we may suspect something is wrong. It may be a technical issue that requires a machine overhaul. This allows us to identify and solve the problem, leading to increased efficiency and reduced production costs.
Can you give specific examples of such simple improvements?
Let’s say that machines in a factory tend to stop frequently due to fuses tripping. In such a situation, we can install a simple system for maintenance engineers that will force the operator to mark the cause of the fault, for example, “fuse problem.” This gives us precise data on how often a problem occurs on individual machines.
An example of a simple solution we have implemented for one of our clients is a system of pop-ups, a type of notification that reminds operators to clean the machines. In one of the factories we work with, operators must clean the pipes that remove dust from laser processing every two hours. We introduced a system that automatically displays a notification on the operator’s panel, reminding them of this obligation. This simple solution has significantly increased the efficiency of the machines and reduced the number of failures.
What is the second step in digitization after collecting basic production data?
Data collection is the first step. Simply having the data already allows us to conclude, but using the data’s potential requires its visualization and more in-depth analysis. Visualization allows for easier understanding of patterns and faster decision-making. We can identify areas that require additional monitoring. If one machine is operating less efficiently than others, we can install additional sensors to monitor its operation more closely.
When we started collecting data on the condition of machines and the number of units produced in one of our clients’ factories, we quickly identified that specific machines had more frequent downtime. We introduced a system that automatically collects data on the causes of failures and suggests ways to repair them. This allowed us to reduce the number of downtimes and increase production efficiency.
Even simple data analysis allows us to identify recurring problems. For example, one machine may have much more frequent failures due to fuses than others. This may suggest that the machine has a hidden defect and requires a more thorough inspection or repair. It may also indicate the need to replace parts that are more prone to failure.
What types of sensors are worth installing in production, and what data can they collect?
We can install a range of sensors, including cameras, as well as those for measuring motion, temperature, pressure and many other environmental phenomena. They can detect unusual patterns in machine operations, which may signal potential issues or faults. Cameras can analyze the movements of operators and machines, identifying areas where efficiency can be improved.
What are the benefits of more advanced sensor applications on the production line?
Additional sensors allow for more detailed analysis and a better understanding of production processes. They monitor various parameters affecting performance, such as light, heat, motion, moisture, and pressure, which allows for faster response to potential problems and reduced downtime.
Sensors allow you to monitor various parameters of mechanical components, such as pressure level, oil quality, and perhaps humidity. In addition, factories can sensor the electrical system to measure voltage and current and look for unusual fluctuations or exceeding limit values. This will show us if something needs to be fixed.
Do improvements always involve changes in the sensorization of the production line?
It is worth noting that better production sensing often results in simple improvements that have a significant impact on efficiency.
Modifications can be implemented at many levels. Properly installed sensors are key in achieving better production efficiency, as they collect the data necessary to determine required changes. However, they are not the only factor. Equally important areas for improvement include the ergonomics of the work environment, knowledge transfer, and the way data is handled – its analysis, integration, etc.
An excellent example of a simple change to improve work ergonomics we implemented for one of our clients is modifying the user interface (UI) on the production line control panel. Thanks to our changes, the operator can no longer press three buttons sequentially to start the production process; now, they can initiate it with just one button press. This change saves time and reduces errors.
Another example is reducing the time needed to train new employees, as simpler and more automated systems are easier to master. Saving a few seconds per operation might seem minor, but it can result in substantial gains over a year and with a high volume of operations. In other cases, improvements can lead to a significant reduction in the number of failures, which in turn translates to savings in repair and downtime costs.
Are there specific metrics that can be monitored to evaluate the effectiveness of these improvements?
Certainly, many metrics can be monitored, such as downtime, the number of failures, machine performance, and the time needed to train new employees.
Read also: Top 6 AI-driven Strategies To Improve Productivity in Manufacturing
Let’s move on to the third step. What actions can be taken when applying additional sensors and collecting more data?
After collecting and analyzing more advanced data, the next step is implementing Predictive Maintenance (PdM). This stage can be initiated after an extended period of data collection, typically a year or more, to draw long-term conclusions and predict machine failures before they occur.
What exactly is Predictive Maintenance?
Predictive Maintenance analyzes collected data to predict potential failures and plan maintenance in advance. This allows you to minimize downtime and maximize machine efficiency. For example, if you notice that a machine usually needs a part replaced once a month, but suddenly it starts needing parts replaced more often, that’s a sign of a problem. We can then schedule an inspection of this machine before a more severe failure occurs.
What are the steps to implement Predictive Maintenance?
First, you need to collect data over a longer period to have a solid base for analysis. Then, using this data, you can identify patterns and trends that may indicate upcoming problems. Based on this, the system can automatically generate alerts and reminders for maintenance managers, informing them when maintenance is due.
What are key benefits of implementing Predictive Maintenance?
The main benefits are reducing unexpected downtimes, increasing machine efficiency, and optimizing maintenance costs. For example, we can identify machines requiring more frequent maintenance and take appropriate preventive actions thanks to predictive maintenance. This not only extends the life of the machines but also reduces the costs associated with unplanned repairs.
Can you give an example of implementing Predictive Maintenance in practice?
Sure. After collecting data for a year in one of the factories we worked with, we noticed that one of the machines was starting to show an increase in the number of failures. By analyzing data from vibration and temperature sensors, we realized the problem was the incorrect calibration of one of the parts. We implemented a predictive system that automatically monitored these parameters and sent alerts whenever the values exceeded certain standards. This allowed us to respond quickly and minimize downtime.
What are the long-term benefits of implementing Predictive Maintenance?
The long-term benefits are primarily a better understanding of production processes and the ability to optimize them. Thanks to detailed data, we can make better decisions, minimize downtime and increase efficiency. This also prepares the factory for future challenges and allows you to remain competitive on the market.
Can every factory benefit from these technologies?
Yes, definitely. It is worth starting with small steps and gradually implementing more advanced solutions. Even simple systems, such as automatic reminders for operators or collecting data from PLCs, can bring tangible benefits. It is crucial to start digitizing production and gradually developing it.
Read also: How to Implement Predictive Maintenance Using Machine Learning?
What advice do you have for factory leaders considering digitizing their production processes?
My tip is simple: start small. You don’t have to invest millions in advanced systems right away. It is worth collecting basic data and gradually implementing more advanced solutions. This can significantly improve the efficiency and effectiveness of your production.
At the same time, it is essential to remember that any delay in data collection costs money. Each delay means further months, and often years, before more advanced systems, e.g. predictive systems, can be implemented because such systems must work on large sets of historical data. So, regardless of your long-term factory digitalization plans, it’s always worth considering the simple first step of transmitting and collecting key production data.
Optimize your factory’s efficiency and reduce downtime with NeuroSYS’s digital solutions. Book a free 1-hour consultation to receive expert guidance on analyzing production data, implementing quick optimizations, and preparing for advanced systems like predictive maintenance. Don’t wait – schedule your consultation today and start improving your operations.
