Privacy architecting of GDPR-compliant high-tech systems : the PAGHS methodology

High-tech systems are getting increasingly complex. To provide a function, a system's components continuously interact with each other exchanging massive amount of data. Such information often involves personal data which started new concern on privacy in the European Union. Recently, the General Data Protection Regulation (GDPR) introduced stricter requirements processing personal data. In particular, it enforces the implementation of data protection by design by default in products. Companies that do not demonstrate compliance are liable for up to 4\% of their annual revenue. Consequently, the regulation is dictating new needs for organizations to create privacy-aware products. System architects have to address these needs with the goal of realizing a product's architecture that fit these needs. Because, privacy is a relatively new concern, the lack of supporting guidelines. We addressed the challenge in two steps. First, we investigate how to fill the semantic gap between legal requirements and technological implementation. Second, we structured an iterative, five-steps process which supports communication with stakeholders, and demonstrates compliance via structured documentation. We validated our work in four empirical sessions with system architects to validate the application of PAGHS. Moreover, we collected feedback from privacy experts on the quality of PAGHS outcomes.