A review of product design driven by the integration of digital twin and knowledge graph

Complex product design is facing the dual challenges of semantic deficiency in multi-source heterogeneous data and lag in the dynamic perception of design knowledge; this fragmentation of data and knowledge constrains the efficiency and innovation capability of product design. To this end, the evolutionary patterns of single technological paths are first revealed in this paper: Digital Twin (DT) is transforming from static verification to proactive evolution and human-machine collaboration, while Knowledge Graph (KG) is upgrading towards generative design collaboration by leveraging Large Language Models (LLMs). However, research on the integration of the two is still in its nascent stage, lacking universal guiding methods for the entire design process. To address this gap, a deep integration-driven guidance framework of DT and KG for product design is proposed. This framework innovatively constructs three core operational mechanisms: semantic intermediary mapping, simulation-optimization closed-loop correction, and knowledge-driven self-evolution, proposing a collaborative "reasoning-verification" operation mode with KG as the "brain" and DT as the "perception". Specifically, through a case study on the optimization design of a high-speed train bogie, the effectiveness of this framework in design guidance is validated. Finally, the key challenges faced during the integration process are analyzed, and future research and development directions are explored, aiming to provide theoretical and technical support for the intelligent upgrade of complex product design. • Systematically reviews the construction methods, applications, and limitations of Digital Twin (DT) and Knowledge Graph (KG) in product design. • Proposes an integrated DT-KG framework to drive intelligent product design. • Validates the framework's advantages (cost reduction, improved knowledge reuse) via a high-speed train bogie optimization case. • Analyzes DT-KG fusion challenges and future directions in product design stages.