Competence requirements for AI assistance systems in production

The introduction of AI assistance systems in production is revolutionizing industrial processes and presenting companies with new challenges. In addition to technological adjustments, this development primarily requires a reorientation of employees' skills. This article examines the general and specific skills requirements that are necessary for the successful use of AI systems in production and summarizes scientific findings on the changes in these requirements.

The skills required for dealing with AI systems in production can be divided into three main categories ¹:

1. technical and basic knowledge:

• Professional competence: Employees have the necessary technical knowledge/skills to perform the day-to-day tasks appropriate to their position. Depending on the employee's position, this may also include manual skills, for example.
• Basic digital skills: Employees handle conventional digital media and technology safely and confidently and can work smoothly with common office programs and digital collaboration technologies in particular. In particular, they have sufficient awareness of digital security aspects.
• AI awareness: Employees are aware of the AI systems used in the company and their basic capabilities; this particularly includes knowledge of what AI systems cannot do. They are sensitized to the data that the AI system processes, including possible personal data.

2. development of AI systems and handling of AI systems:

• MMI skills: Employees have skills for targeted handling in human-machine interaction at the current state of the art.
• Basic knowledge of machine learning: Employees know and understand the basics of machine learning, including deep learning and neural networks, and can apply this knowledge in human-machine interaction.
• Skills in programming languages, platforms, frameworks and libraries: Employees are proficient in relevant programming languages such as Python as a basis for machine learning. They are confident in using common platforms such as Amazon Web Services (AWS) and frameworks/libraries such as Sparks or Hadoop (Büchel & Mertens 2021).
• Big data, data science and data analytics: Employees have skills in managing, collecting, compiling, processing and modeling data and analyzing large volumes of structured or unstructured data. Important areas of expertise include advanced mathematics, cryptography, data ethics and data privacy or data mining (Gesellschaft für Informatik 2019).
• Process and system skills: Employees can recognize processes and workflows in the company, think in terms of these processes and workflows and structure their own work behavior in processes and workflows. They are also able to describe, reconstruct and model these processes and other complex issues as systems and, on this basis, make forecasts and develop options for action. In concrete terms, employees realize the specifics of the influence of AI on company processes: They understand the changes brought about by AI and can optimize their own work processes in relation to collaboration with AI.
• Problem-solving skills, resilience: Employees can quickly recognize unexpected situations and difficulties, deal with them and develop suitable solution strategies. This includes in particular the knowledge and, if necessary, the practical ability to intervene in AI-controlled processes.
• Reflection skills: Employees are able to critically interpret and evaluate the information and results of AI systems. They can independently and competently assess when trust in AI systems and the data generated by AI systems is justified.

3.shaping the context of AI:

• Self-competencies: Employees have a sufficient degree of personal responsibility and self-organization. They have the curiosity and willingness to learn and work with machine learning and AI technologies.
• Social and communication skills: Employees can contribute to teams with different compositions. They can work together with colleagues from different professional backgrounds and with different levels of experience and expertise. In contact with customers and users of AI systems, employees can explain the special features of the use of AI systems for their respective area of responsibility.
• (Personnel) management, leadership skills, change management: Employees with leadership responsibility can organize a team, coordinate and delegate tasks (bundles). They can communicate the potential and limitations of AI, allay fears and activate further training potential. When integrating AI systems into company processes, they can formulate reasonable goals and thus help shape the change process.
• Decision-making skills: Employees know their responsibilities and are able to make reliable, well-considered decisions within the scope of their responsibilities.
• Adaptability, transfer: Employees are able to adapt to the opportunities and challenges presented by AI and adjust their working methods accordingly.

Information technology skills and domain-specific expertise are equally important. However, employees who have mastered both areas of expertise are rarely available ². Social-communicative skills and ethical values are therefore becoming increasingly important.

Specific skills requirements

Various roles with specific skills are required for the introduction of AI applications in production ³:

1. Company management, project sponsorship:
   • Strategic focus, company-wide perspective, confidence in AI technologies, decisiveness and foresight.
2. Project management / project expertise:
   • Project and people management, leadership skills, ability to coordinate interdisciplinary teams.
3. Technology experts, automation experts:
   • Problem-solving skills, resilience, ability to reflect, knowledge of signal processing and automation technologies.
4. Data scientists, data engineers:
   • Data selection, preparation, analysis and interpretation, teamwork and communication skills.
5. AI experts/ ML experts:
   • Sound knowledge of mathematics, statistics, machine learning and data management.
6. MLOps engineers, IT security experts:
   • Fundamentals of software engineering, process and system expertise, knowledge of IT security.
7. Process managers, maintenance staff:
   • Professional competence, in-depth process knowledge, domain and operator knowledge, self-competence in systems engineering.
8. Quality managers:
   • Knowledge of operational quality specifications and QM systems, ISO standards, tolerance management.
9. Safety specialists, occupational health and safety experts:
   • Expertise in ethical, legal and social implications (ELSI), process and system competence.
10. Plant/process operators:
    • Domain knowledge, practical experience, ability to interact with AI systems.
11. Works council:
    • Mediation skills, mouthpiece between workforce and management, understanding of ELSI issues.
12. Interaction designers:
    • MMI expertise, design skills, focus on usability and user experience.
13. Personnel developers/ change managers:
    • Communication management, skills development, promoting transparency, open communication.
14. System developers:
    • Programming skills, understanding of company IT architecture.

Everyone involved should have basic digital knowledge, be communicative, show adaptability, be creative and open to new ideas ³.

1. Definition of (job) roles and responsibilities in the context of AI: The first step is to define the responsibilities of the respective (job) roles and their interfaces to other areas in detail. This is done along the core tasks using suitable methods such as the RACI method. It is essential not only to define the areas of responsibility, but also to name the activities that lie outside the responsibility of a role in order to prevent competence profiles from being unnecessarily expanded. This process requires a deep understanding of company structures and processes.
2. Assignment of tasks in the changed division of labor between humans and AI: Based on the defined roles, the core and detailed tasks are assigned to the respective roles. This is only possible through close cooperation with the departments concerned, as the daily practice and professional insight of the employees must be included. The result is a tabular overview of the (job) roles with associated tasks, which serves as the basis for assigning specific AI competencies.
3. Derivation and definition of specific AI competencies for task fulfillment: The competencies required to fulfill the tasks are divided into professional, methodological, social and personal competencies. A competence profile comprises the knowledge and skills ("can"), the responsibility ("may") and the motivation ("want") to fulfill a task. These components must be taken into account in competence management in order to create realistic competence profiles. Competencies are assigned according to the tasks of a (job) role.
4. Definition of competence profiles and determination of target profiles: Based on the tasks of the (job) roles, competency profiles are created that define the necessary competencies and their characteristics. This step requires critical consideration to ensure that only the relevant skills are taken into account. The target profile, often presented in the form of a network profile, shows the level of competence required for a role.
5. Competency needs analysis and individual assessment: In this step, employees are assigned to the competency profiles created and their current level of competency is assessed. This can be done through observation by line managers or through a cooperative method in which line managers and employees develop and discuss the current profile together. This method promotes motivation and personal responsibility. The analysis provides the basis for planning further training measures.
6. Determining suitable further training measures for AI skills development: Based on the skills needs analysis, specific further training measures are determined to close the identified gaps. Learning opportunities that combine formal and informal learning are particularly effective here. Digital learning technologies increase flexibility and enable practical implementation of what has been learned. Skills development is evaluated and adjusted in employee appraisals to ensure sustainable learning success.

This structured process of skills management ensures that employees are able to cope with the new requirements of AI assistance systems and continue to develop their skills.

Changes to the skills requirements

The use of AI assistance systems leads to significant changes in skills requirements:

• Increase in job complexity and skill variety: the complexity of tasks increases, particularly in information-processing activities, which requires enhanced skills ⁵.
• Possible reduction in autonomy: AI systems can restrict the autonomy of employees in low- and medium-skilled activities ⁵.
• Shift in task focus: In assembly, the focus shifts from cognitive to manual non-routine tasks ⁵.

Skills development

Targeted skills development is crucial for the successful implementation of AI assistance systems ^{6 3}:

• Strategic integration: Skills development should be an integral part of the corporate strategy ⁴.
• Practice-oriented training: "On-the-job training" and "in-house seminars" promote direct application and facilitate implementation in everyday working life ^{3 8}.
• Combination of formats: Digital learning opportunities should be supplemented by experience-oriented face-to-face events in order to meet different learning needs ^{3 8}.
• Inclusion of all employee groups: Low-skilled employees in particular should be included in skills development measures to prepare all employees for change ^{3 8}.

The successful introduction of AI assistance systems in production therefore requires comprehensive skills management that takes into account technical, specialist and interdisciplinary skills and involves all employee groups ^{7 3}.

Conclusion

The introduction of AI assistance systems in production requires more than just technological adaptations. It requires comprehensive skills management that takes into account technical, specialist and interdisciplinary skills. The ability to combine domain-specific knowledge with AI skills and to react flexibly to new requirements is of central importance here ^{7 3}. Companies should therefore invest in the continuous training of their employees and promote a culture of lifelong learning in order to fully exploit the potential of AI in production.

Sources/Footnotes

1. Learning Systems Platform. Skills development for AI - needs and solutions for education and training. Retrieved from https://www.plattform-lernende-systeme.de/files/Downloads/Publikationen/AG2_WP_Kompetenzentwicklung_KI.pdf
2. Platform Learning Systems. AI and work - competencies. Retrieved from https://www.plattform-lernende-systeme.de/schwerpunktthemen/ki-und-arbeit/kompetenzen.html
3. Fraunhofer IAO. Human-centered AI applications in production. Retrieved from https://www.ki-fortschrittszentrum.de/content/dam/iao/ki-fortschrittszentrum/documents/studien/Menschzentrierte-KI-Anwendungen-in-der-Produktion.pdf
4. Platform Learning Systems. AI Competence Development in Office and Production Work. Retrieved from https://www.plattform-lernende-systeme.de/files/Downloads/Publikationen_EN/AG2_WP_AI_competence_GB.pdf
5. Dombrowski, U., & Wagner, T. Impact of Artificial Intelligence on Work in Assembly Systems. Journal of Intelligent Manufacturing, 2023. Retrieved from https://link.springer.com/article/10.1007/s10845-023-02086-4
6. Rossi, A., et al. Artificial Intelligence in Industry: A Review on Uptake and Competencies. 2022. Abgerufen von https://re.public.polimi.it/retrieve/e0c31c12-685b-4599-e053-1705fe0aef77/SSRN-id4072671.pdf
7. Handelsblatt Live. AI in the workplace: Which skills are in demand now. Retrieved from https://live.handelsblatt.com/ki-am-arbeitsplatz-welche-kompetenzen-jetzt-gefragt-sind/
8. Platform Learning Systems. Skills development for AI - needs and solutions for education and training. Retrieved from https://www.plattform-lernende-systeme.de/files/Downloads/Publikationen/AG2_WP_Projektbericht_Kompetenzentwicklung_KI.pdf
9. Stifterverband and McKinsey: Future Skills Framework, 2021 Retrieved from https://future-skills.net/framework