Home » “AI is becoming more and more a team game”

“AI is becoming more and more a team game”

by admin
“AI is becoming more and more a team game”

AI is also increasingly being used in production environments to relieve employees and address the shortage of skilled workers. You can read about the potential use of AI, what the major challenges are and why AI ethics are essential in the interview with Jens Beyer from LAVRIO.solutions from the CyberForum network.

Von Ana López

What specific applications and benefits does AI offer in today’s production environment? How can AI technologies be used?

There are diverse applications of AI that bring revolutionary changes. For example, for several years now we have seen quality controls supported by AI methods that make work processes much more ergonomic and efficient. As production environments become ever faster, this relief for employees is clearly noticeable. In addition to quality controls, real-time monitoring of the entire process is also a profitable AI application. AI can detect anomalies in the process and suggest corrective actions. Likewise, methods for training new employees or seasonal workers are an area in which AI solves many of the challenges faced by manufacturing companies. For example, the AI ​​systems provide targeted support to new employees at work and provide suggestions and tips for the next work steps.

Keyword “predictive maintenance” – AI is already being increasingly used in the maintenance area…

In the area of ​​maintenance, there are various developments that improve the longevity of machines and minimize maintenance downtime. In addition to dashboards with concrete AI-generated recommendations for the next maintenance steps, “predictive maintenance” is of course also an AI system that is now used in many places. This predicts which part of a machine might fail and when it should be replaced.

What challenges and risks are associated with the use of AI in production processes and how can they be minimized?

Technical challenges lie, for example, in data quality: AI systems are heavily dependent on the existing training data. Inaccurate, incomplete or unreliable data can quickly lead to errors. Integration into existing systems is also technically challenging. Introducing AI systems into production environments can be complex, especially when legacy machines and systems are involved.
There is also a lack of qualified specialists who are experienced in both AI technology and production. However, these are needed for the introduction and maintenance of AI systems.
Social and organizational aspects also play a role. A certain level of skepticism towards AI systems can lead to delays in the introduction and circumvention of the systems or even sabotage by the skeptics.

See also  Cyberpunk 2077's new stunning lighting mod simulates real-world cinematic shots -

How can you increase employee acceptance of AI?

It is important to openly engage everyone involved in the process right from the start, train them and involve them strongly in the introduction process. Transparency of the AI ​​systems, which we can achieve through technical measures (keyword “explainable AI”, or “explainable AI, XAI”) helps to increase acceptance. The focus on hybrid systems in which AI suggestions are paired with human expertise also significantly helps to build acceptance.
Training courses and workshops should be offered throughout the introductory phase and at the beginning of the productive phase in order to keep a clear eye on the organizational and technical challenges. This means you can react quickly when a new challenge arises. This is an important topic that we continually address in research projects such as KARL (“AI in work and learning in the Karlsruhe region”).

How to ensure that AI is used responsibly in production ethics?

The ethical implications of AI are of key concern to us, particularly in areas where they have a direct impact on human workplaces, security and privacy. In the context of production, these include employee safety, data protection and process quality.
There are several starting points for using AI systems responsibly. On the one hand, the AI ​​systems should be designed transparently so that the AI’s decision-making processes are comprehensible and explainable. This is particularly important in areas where AI decisions have a direct impact on workers. On the other hand, the final decisions remain with people. Ideally, AI offers very good decision-making support.

Employee participation is extremely important, especially from an ethical perspective. Only well-trained employees who understand AI technology can assess the ethical implications.

What about data protection?

Data protection should also be taken into account from the outset. When collecting and processing data, we must always check to what extent privacy and labor law framework conditions are being complied with.

See also  Without a pitch deck to millions from Lightspeed and Sequoia

What should you pay particular attention to with multinational companies?

For companies with locations around the world, a certain cultural sensitivity is of course also part of the responsible use of AI. Depending on the region, the differences that exist with regard to AI must be taken into account.

How important are ethical standards for the use of AI?

It is essential for companies to develop clear ethical guidelines for the use of AI. We see pioneers here in the Karlsruhe region who are working on such principles in their digital laboratories. They cannot be rigid and final yet, as AI technology and its areas of application are continuously developing. But a foundation can and must be created now. This increases the trust of employees and is a good basis for this leap in productivity.

What skills will employees in industry need in the future?

The main aim of the AI ​​projects that we are currently supporting and seeing are to use recommendations from the AI ​​systems to relieve the burden on human workers, automate unergonomic process steps, reduce speed and reduce stress. Of course, this means that employees have to be open to new things. We need contacts who know their processes, but on the other hand accept improvements and new processes.
It is therefore important that they acquire basic skills or knowledge in data analysis, machine learning and robotics. Only then can they critically question the AI ​​systems and identify emerging problems more quickly.

What are the best practices?

Fortunately, there are now many concrete examples available. For example, in the “KARL” research project there are some exciting use cases that clearly illustrate the advantages in production. An assistance system for the assembly of electric motors was developed here. Various AI models were used to support the overall system, for example hand movement detection using sensors and the company’s AI models Kinemic or direct support for the workers Optimums “Smart Klaus”, who monitors the production process with a camera and provides direct information on the next step. This project clearly aims to increase the well-being of employees by reducing the potential for errors and stress. At the same time, it also helps to minimize a “skill gap” because all workers receive this support.

See also  This AI camera takes photos without shooting

The big fear is losing decision-making power…

With these AI systems, decision-making authority remains with people. A production-related use case, for example, is the one in collaboration with INIT emerging support for transport managers in local public transport. Here we develop as LAVRIO.solutions a model to support employees in a time-critical situation. The traffic dispatchers not only receive a recommendation from the AI ​​about what they should do next, but also a direct indication of why the AI ​​is making this recommendation. This allows them to decide whether to follow this recommendation or whether the AI ​​has not taken all the decisive factors into account.

So humans and AI as mutually supporting partners?

Absolutely. We see from these examples that the focus is on supporting people. At the same time, these use cases are successful because the users and everyone interested in the process in the respective companies are strongly involved from the outset. AI is becoming more and more a team game in which the companies that derive the most benefit from it are those who implement the projects technically flawlessly and at the same time tackle the organizational challenges from the outset.

Jens Beyer is a data scientist and AI consultant at Lavrio.solutions GmbH in Karlsruhe. His focus is on the rapid, interactive development of AI systems and the conception of AI modules for existing software applications. He accompanies software companies on their path to AI and advises companies on AI implementation projects.

Do you want to know more about the topic, go deeper or further educate yourself on AI?
At the CyberForum Academy you will be equipped for your digital future with intelligent skills and methods. View digitalization holistically, learn the basics of AI and find out how your AI project will be a success.

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Privacy & Cookies Policy