This paper presents the methodological foundation for ecological sustainability assessment in early product development using knowledge graphs (KGs). Conventional life cycle assessment (LCA) methods are often impractical at this stage due to incomplete data, various data sources, and divergent stakeholder objectives. The proposed approach outlines an evolving framework that aims to enable dynamic data integration, systematic handling of uncertainty, and individualized assessment of sustainability indicators. First exploratory applications illustrate the feasibility of AI-supported KG functions, such as linking product entities with external CO₂ databases and generating synthetic data. While the methodology is still under development and has not yet been validated, these preliminary results highlight the potential of KG-based methods to provide transparent and adaptable sustainability assessments in the early design phases.
Svenja Hauck (Wed,) studied this question.