SID: Secure data sharing for Industrial Digitalization

Purpose and goal:

The digital transformation of industry increases data collection and the need to share data between processes and stakeholders. More data collection and combined data flows from multiple systems open up great opportunities for data analysis and machine learning. But there is an obstacle.

Data is often sensitive to share between different stakeholders and between companies and customers. This prevents the use of data. The project objective is to enable Swedish industry to use privacy-preserving techniques to share and analyze sensitive data and avoid the obstacles that exist today. Expected results and effects:

Privacy-preserving techniques provide cybersecurity in the form of avoiding unwanted consequences - leakage of sensitive data - should an incident occur. The project will deliver public reports, practical guidelines and scientific publications to share lessons learned. A collection of case studies, with associated analysis and identification of pitfalls, will serve as a practical guide for the application of privacy-preserving data analysis.

This will make it possible to use sensitive data for analysis and models to a greater extent, which promotes data-driven development. Approach and implementation:

The project will be performed in close collaboration between the project partners RISE, Sensative, Bron Innovation and industrial partners in the project reference group. The project will carry out case studies with real-world data sets and conduct privacy-, utility- and performance analysis of differential privacy, federated learning and also federated learning combined with differential privacy and homomorphic encryption.

The project will arrange workshops with stakeholders from Swedish industry to discuss real-world use cases and disseminate the project results.