Künstliche Intelligenz als Innovationstreiber: Potenziale in der Qualitätsprüfung der Automobilindustrie

Abstract

Artificial intelligence (AI) is considered a key driver of innovation in quality testing in the automotive industry. In view of increasing demands for efficiency, precision and documentation, traditional testing methods such as manual visual inspections and random sampling are increasingly reaching their limits. They are error-prone, cost-intensive and often unable to adequately reflect the growing complexity of modern vehicle components. However, standards such as ISO 9001, VDA Volume 5 and Volume 5.3 require comprehensive monitoring and documentation of quality assurance and systems, which further increases the need for innovative solutions. The aim of this thesis is to analyse the potential of AI technologies, in particular machine learning and deep learning, for the automation and optimisation of quality testing. The efficiency and accuracy of AI-supported testing methods are compared with traditional methods, and the challenges of successful implementation are examined. The thesis answers three central research questions: Which AI technologies are particularly suitable for quality testing? How do the results differ from those of traditional testing methods? And what technical, organisational and ethical hurdles exist when integrating AI into existing processes? The methodological approach is based on a hermeneutic-interpretative analysis of current scientific literature, supplemented by case studies and practical examples. The results show that AI-based systems offer a significantly higher error detection rate and inspection speed. They enable 100% inline quality control and reduce both personnel costs and expenses. At the same time, they ensure continuous improvement by learning from historical data and dynamically adapting to new error patterns. However, challenges remain: the integration of AI requires high data quality, specific technical infrastructure and qualified personnel. Acceptance issues, a lack of transparency (keyword: ‘black box’) and ethical questions such as fairness and data protection must be addressed. The work emphasises that explainable AI (XAI) and transparent decision-making processes are particularly crucial for trust and acceptance in the industry. In summary, the analysis shows that AI has the potential to fundamentally transform quality testing in the automotive industry. It offers solutions to existing weaknesses in traditional methods and opens up new avenues for greater efficiency, precision and cost-effectiveness. In the future, AIsupported quality control is expected to become standard, with companies investing in data management, training and ethical frameworks.

Date of Award	2025
Original language	German (Austria)
Supervisor	Josef Wolfartsberger (Supervisor)