One of the biggest obstacles to data collaboration in maintenance to date is the concern about protecting sensitive information. In particular, wear models for critical components often require context and process data in addition to individual component data, which in many cases is classified as confidential. This often prevents collaboration with experts so that monitoring or wear prediction is not possible.

To overcome this challenge, we are introducing a range of innovative technologies that make it possible to share and utilise sensitive data without compromising confidentiality:
– Homomorphic Encryption allows machine data to be analysed with third-party models without losing confidentiality
– Federated Learning enables machine operators to jointly create models without revealing their sensitive data
– Differential privacy can be used for secure and anonymous benchmarking

Learn more about the possibilities of data collaboration while maintaining confidentiality and discuss with us how your process data can be utilised or existing wear models can be protected.
Visit us at our stand A09 in hall 4!