Digital Twin: Ten Fundamental Questions and Insights for Researchers, Decision‑Makers, and Practitioners

The paper examines ten key questions about digital twins to stimulate discussion and provide reference for researchers, decision‑makers, and practitioners. It begins with a definition of digital twins, identifies who is interested in the technology, and compares the research activity in China, the United States, and Germany.

It then explores the relationship between digital twins and smart manufacturing, emphasizing how digital twins enable information‑physical integration, real‑time feedback loops, and intelligent decision‑making across the product lifecycle.

Five dimensions of digital twins are analyzed: model dimension (multiscale geometric, physical, behavioral, and rule‑based models), data dimension (large‑scale, multi‑source, real‑time data and its fusion), connection dimension (cross‑protocol, cross‑platform, bidirectional interaction), service/function dimension (simulation, monitoring, optimization, fault prediction, etc.), and physical dimension (the inseparable link between the virtual model and the physical entity).

The article discusses how digital twins intersect with emerging “New IT” technologies such as IoT, AR/VR/MR, edge computing, cloud computing, 5G, big data, blockchain, and artificial intelligence, illustrating that these technologies are essential enablers for real‑time, high‑fidelity digital twins.

Scientific challenges are identified across the five dimensions, including intelligent sensing of heterogeneous physical assets, dynamic high‑fidelity modeling, massive data acquisition and fusion, real‑time cross‑platform connectivity, and delivering tailored services for diverse applications.

Standardization needs are highlighted, noting ongoing work by ISO/TC184, IEEE 2806, and ISO/IEC committees, while stressing that a comprehensive digital‑twin standard framework (common, key‑technology, tool/platform, evaluation, security, and application standards) is still lacking.

Finally, the paper reviews commercial tools and platforms (e.g., MATLAB Simulink, ANSYS TwinBuilder, Microsoft Azure, Dassault Systèmes 3DEXPERIENCE) and points out their limitations in covering all functional dimensions, openness, and interoperability, concluding that future digital‑twin adoption requires integrated, open, and compatible commercial solutions.