Data-Driven Approach To Condition Monitoring To Improve The Service Business – Henrik Pedersen

Over the last two years, the condition monitoring system at Siemens Wind Power A/S has transitioned from a purely physics based monitoring system to a mixed physics based and data-driven system. By utilizing data fusion and machine learning to detect and classify defects and abnormalities in data, the system is now able detect damages at a stage where a human expert is challenged to verify the finding. This has significantly improved the performance and stability of the system and allowed the company to develop new digital products.


  • Challenges in incorporating data science in a large company not born in the digital era.
  • Why a data-approach is a good supplement/replacement for a purely physics based system.
  • How can a traditional service business benefit from data science and machine learning?

Add comment