The Culture Clash of AI Adoption in Lean Quality Management. Resolving the Tensions at Siemens Electronics Works Amberg

Abstract

Artificial intelligence (AI) brings great potential for manufacturers, but clashes with the established culture due to the unexplainable and opaque nature of the solutions it provides. Having little experience in AI and machine learning (ML), most manufacturing leaders experience barriers implementing AI. This is especially true in lean environments, since these are often nearly perfected and hence intolerant of failure. Cultural barriers to adoption abound, even if the quality of the AI is assured. Deeply grounded in the case of Siemens' Digital Industries division, this 5-year practitioner research investigates AI adoption in lean quality management. Collaborating with key executives involved in the case and fellow researchers, we investigate the challenges and solutions of AI adoption in lean quality management. We identify three distinct paradoxes: (1) Human-driven versus AI-driven analysis, decisions and actions, (2) transparency versus opacity and (3) specification-driven versus discovery-driven processes to achieve quality. Rooted in Siemens' experience, we derive four strategies for overcoming these paradoxes. Our findings have important implications for practice, as we present clear and realistic steps and guidance for effective AI integration into lean quality management environments. Our results contribute to the discourse on AI adoption, pointing to the importance of recognising potential cultural clashes as well as strategies for overcoming these which go beyond the technology itself.