The use of data-driven techniques can support the trade-off between different PSS design concepts by allowing dynamic indexing and retrieval of information to detect correlations and emerging trends. The chapter presents a data-driven approach exploiting historical PSS operational data and Computer Aided Design (CAD) models to populate value models for alternative design configurations inside the feasible design space. The approach encompasses the use of the Design of Experiment techniques and machine learning (ML) to create surrogate models of value for each potential PSS design configuration. The application of the approach is described through two case studies run in collaboration with a tier-one aero-engine sub-system manufacturer and an original equipment manufacturer in the construction equipment field.