The Human Touch in Modern Manufacturing

Without good tools, production as a concept or object cannot exist. If you mentioned the word "production" to someone unfamiliar with the field, they would probably associate it with some tool used in the production process. Even the most miniature workshops have tools to help employees manufacture parts. So, tools are a necessity. Of course, in this case, I am talking about physical tools. Let's start with an example, such as a welding machine. If a manufacturing company deals with welding steel structures, this is the primary tool in that production. Without a welding machine, such a company can hardly survive. Of course, there are options for the company to only deal with the planning and drawing of structures, have another company manufacture the structure, and yet another to hire for installations and assembly. Still, again, in this entire process, the welding machine is essential. It is only possible to make a welded structure with a welding machine. So, in this case, the tool, namely the welding machine, is a critical component that companies must carefully maintain and invest in newer, better welding tools. Thus, each type of production has tools in the production process that are critical in one way or another, which is necessary to carry out the production process. Therefore, these tools, which can also be machines, robots, or, ultimately, lines, are essential for companies to transform raw materials into a final product.

On the other hand, the most advanced and successful manufacturing companies realize that their competitive advantage lies in always being a step ahead of the competition in introducing new tools and innovations. Suppose a new tool can save a certain amount of time. In that case, it means that the company can produce a particular piece faster than the competition and consequently bring it to the market at a lower price and increase its market share, or sell it under the same conditions and thus secure a more significant profit. Reinvesting this profit in the production process is advisable for achieving constant growth.

TIP: The right tools are vital for survival and success in manufacturing. Beyond just functionality, strategic tool selection and innovation provide companies with a competitive edge, enabling faster production, cost efficiency, and market leadership.

Crafting Tomorrow: Humans Meet Automation

However, even though these tools are crucial for business success and production, humans operate such tools. Even with introducing the most modern tools in the production process, the human is still the critical component in the equation. It may change in the future. But rather than wondering if it will change, we should ask when. Until now, creating a new lathe with an existing one has been a standard; however, in the future, tools will evolve to configure and create new tools autonomously. That is, tool settings, defining the tool based on past experiences, and setting goals.

Companies that have traditionally depended on the experience and skill of machine operators to set parameters will soon see these settings adjusted automatically by machine learning. This adjustment will be based on the product's specific requirements, whether with a machine, welding apparatus, or any other tool. And the critical component in preparing these innovations and the most modern tools is artificial intelligence. Machine learning is so revolutionary because of this ability to learn from past experiences. Because, in some way, it eliminates the component of man, but do not worry, not entirely.

TIP: Embrace the future where AI-led automation complements human skill in manufacturing, evolving from manual tool settings to intelligent, experience-based machine configuration for enhanced efficiency and precision.

Mastering the Weld: Skill Over Chance in Manufacturing

Suppose we go back to our welding machine. In the case that a person picks up the welding gun and welds, we can easily conclude that the more skilled the person is in welding, or the better the artist who can assemble the product using the welding machine, the nicer the final product is created and not just the product, the weld itself will also be visually more excellent. You probably also agree that the weld will also be of better quality. As you may have noticed or perhaps even welded yourself and observed, the more experienced welders make an excellent and consistent weld. In contrast, those who are beginners or less experienced make a somewhat distorted weld. We could also discuss the hardness of the weld and its quality. But for our topic, the first group must make the weld as it should be from all quality aspects, while the others are welding more like a game of chance.

If the production plant operated as a casino, leaving everything to chance, it would probably be closed quickly. However, we can observe that despite standards, procedures, high automation, and advanced planning, activities still need improvement. If you asked your wife if production plants are rigid with many prescribed procedures, she would probably answer affirmatively, especially if she comes from a field where interpersonal relationships are at the forefront. The more your wife becomes familiar with the production environment and its organized procedures, the quicker she'll realize that many activities still depend on the individual discretion of the people involved. The higher she would move up the hierarchical ladder, the more such activities she would recognize. And it is in these activities that human judgment is crucial. If the person involved in such activities is experienced and supported by some knowledge, the decision will likely bring a positive result to the company. Therefore, the result is obvious if the more experienced and less experienced employees have the same tool or have to use the same welding machine. In this case, we can easily assume who will make a better decision and ultimately create a better weld.

TIP: In manufacturing, the quality of craft, especially in welding, hinges not just on the tools used but significantly on the skill and judgment of the individual. Advanced tools aid, but the nuanced expertise of seasoned professionals ensures excellence over chance in every product crafted.

AI in Welding: Bridging Skill Gaps

But what if we now add our miraculous ingredient, which currently seems to solve all our problems, not just in production but also those that humanity is now facing? Of course, we are talking about artificial intelligence. Imagine those two employees, one an experienced welder and the other a beginner for whom this is even the first job after finishing high school. Today, they have a welding machine available, the market's most modern welding machine. This welding machine has an integrated solution based on machine learning that helps improve the weld quality. What are the chances for the beginner to outperform the more experienced welder now?

In the welding process, machine learning solutions monitor all parameters produced by the welding machine. It also monitors all aspects of the process, from temperature to the quality of the weld itself, with the help of machine vision. All the gathered data leads to self-regulation, where machine learning algorithms adjust the welding machine's parameters in real time as the person performs the welding. This solution constantly adjusts the current, intensity, and wire feeding to ensure the best possible weld quality. Even in this case, the more experienced welder would create a better weld. He would make a better weld mainly because he has seen countless examples of excellent welds and knows what good and bad weld means. However, the less experienced welder will make far fewer mistakes, especially since he will make a significantly better weld than a welding machine that does not contain such a solution. However, the judgment of whether such a weld is of quality remains in the hands of the person holding the welding gun because he can still create better welding conditions by adjusting the distance of the welding gun or the speed of moving the welding gun during welding.

TIP: AI in welding narrows the skill gap, equipping novices and experts with enhanced capabilities. While experience still leads in quality, AI's real-time adjustments aid in reducing errors, proving its value in elevating overall craftsmanship.

Redefining Expertise: Humans and AI in Automation

But what happens if we want to eliminate the human from the active welding process? Let's say we automate the processes, and instead of a person, a robot holds the gun. Even in this case, a human in the background ultimately defined the process, programmed the robot, set up machine learning for welding monitoring, and so on. They have effectively replaced the person who actively engages in tasks like welding. However, the knowledge and experience of the people involved in the process are still vital in ensuring its quality. Ultimately, it is the human who sets the standards.

Let's say we now look at cooking. We have two people. One who has a passion for cooking and has been honing his culinary skills for years. And another person for whom food does not represent something important in the world does not cook and is satisfied with some simple tastes. He has not even developed taste buds to detect something more flavorful. If we give these two people a robot with the most modern AI for meal preparation, the first one will bring a tastier and more beautifully arranged dish to the table we are sitting at. Because the first person has set higher standards, has a more developed taste, and is more experienced in preparing food. It is where the charm lies: AI and ML solutions are just tools that can help us make our products, processes, and, ultimately, companies even better than they otherwise are.

Therefore, the more experienced welder will make an even better weld and, consequently, a better product. If we translate this to the level of companies, good companies will become even better. Those that are now bad will be less bad compared to the better ones that use the same solution. Of course, this is challenging to assess, but let's say this is the case. However, we can already predict that those companies that are now worse have the potential to become better than those who do not use the most modern tools. It will simply not be possible to compete with only classic and traditional tools. Indeed, we must remember that it is necessary to meet some prerequisites for companies to use such solutions. One of the critical conditions is data.

TIP: In the evolving automation landscape, AI reinforces the vital role of human expertise, from programming robots to culinary artistry. This synergy enhances product quality and business competitiveness, highlighting that human judgment and standards remain indispensable even in an automated world.

Data: The Unseen Powerhouse in Machine Learning

We've discussed the tools and how important the human element is in the process, but there's a third component we must pay attention to data. Data are as crucial as the fuel that powers a machine. In most cases, we're talking about electricity, and if there's a power outage due to a snowstorm, many production processes will halt. The same holds for data. If a machine learning-based solution doesn't receive data, it will stop functioning. But this is true for all software in general. If you're watching digital TV and the data transmission stops, so does your viewing experience. If an ERP system doesn't receive data that someone from another department needs to enter for you to initiate a transaction manually, your work will come to a standstill. This is because creating the transaction requires data that has been carefully and thoroughly updated.

However, with machine learning, it's even more crucial to have a variety of data through which the system can sense the current environment in which it operates and information about past events. Therefore, besides having a robot equipped with a welding machine and machine vision for detecting position, orientation, temperature, and other parameters, as well as other sensors for recording electrical current, voltage strength, and wire movement, it's also essential that such a solution has something fundamental, which might seem obvious to us: understanding what constitutes a good or bad piece. Based on experiences and events, the system continuously learns and recognizes good and bad pieces. If it's an even more advanced solution that adjusts machine parameters, having access to data becomes even more critical.

This means that for any solution involving machine learning, it is crucial to have data available from the past. Otherwise, it will take time for the system to learn what all the parameters mean, good and bad, and other events, exceptions, and subsequent settings. And these events need to occur. It's unlikely that all tasks performed at a workstation will occur within one week. It's also improbable that all seasonal fluctuations will occur within one week, significantly if natural environmental factors like temperature, humidity, and light affect the process. At lower temperatures in the factory, it's necessary to use larger quantities of an additive. And all these details, which may seem obvious, must be recorded and learned by the system. But if you already have these data, the solution will "learn" faster, or experts preparing such a solution will more easily create a functioning system, significantly shortening the path to final implementation and consequently to the first results. Just as it's essential to launch a new product on the market as soon as possible, the company needs to start reaping the benefits of such a solution as quickly as possible.

TIP: Data is the lifeblood of machine learning, as vital as electricity to machinery. Its role in training AI systems and guiding decisions is crucial, making collecting and understanding quality data a foundational step for effective and efficient machine learning applications in any field.

From Mechanics to Code: The Evolution of Manufacturing Expertise

In the future, we are likely to witness an exciting battle. If mechanical engineering and electrical engineering have been necessary for product manufacturing until now, the field of software engineering, or IT, will increasingly come to the forefront. To implement advanced solutions based on machine learning, companies will need staff open to using and implementing such solutions, and employees must see the advantages and benefits. All too often, we can detect skepticism among people when using new technologies, especially in areas where we feel competent. The reason lies in the fear that such a solution might replace us, which is an entirely justified fear given the current advancement of technology. On the other hand, if it means that we as a society have to adapt, then so be it.

Humans are not designed to perform repetitive tasks continuously, and machine capabilities outmatch our ability to handle large volumes of data. On the other hand, machine equipment cannot replace human contact. Companies could focus more on their employees, which might be all right.

TIP: The future of manufacturing melds mechanical and electrical engineering with IT, demanding adaptability and openness to machine learning. This shift, while challenging, offers a new balance where technology enhances human work without replacing the invaluable human element in the industry.

GIVEAWAY: Exploring modern manufacturing reveals the interplay of essential tools, human skill, and now, AI's transformative role. As technology evolves, moving from manual tool operation to AI-driven autonomy, the fusion of human insight with advanced tools becomes crucial for efficiency and innovation. In welding, for instance, AI augments the welder's skill, enhancing the craft rather than replacing it. This synergy across manufacturing sectors highlights combining human expertise and artificial intelligence as key to future production.

Data, the new lifeblood of machine learning, drives decisions and automates processes, necessitating a deeper understanding and utilization by companies. The shift from traditional engineering to software engineering in manufacturing requires embracing change and overcoming skepticism. It's about achieving a balance where technology complements human abilities, leading to a more efficient and innovative industry centered around human capabilities.