Digital twin, as a new industrial technology, provides great opportunities in various stages of product development. Product redesign is widely required in the process of product improvement, which is greatly depends on the functional analysis of product. Although traditional functional analysis can identify product design problems, the analyzed information is extremely detailed and verbose, which hinders the opportunity of product innovation. To expand the solution space for improving the innovation chance and ensuring solution quality of the product in the physical space, a digital twin is introduced in the redesign process. This study proposes a product redesign method using the functional backtrack obtained from a relational function model (RFM) to the hierarchical function model (HFM) with the digital twin. Based on a selected target product, the proposed method constructs the product RFM (sub-field) that originates from the reverse fishbone and relationships between components. Related parameters of components are obtained. A digital twin entity is built using the RFM (sub-field) and parameters based on the target physical product, and functions are extracted in the form of “verb + noun.” The RFM is formed considering four relations between functions. Furthermore, functions in the RFM are divided into various levels using the Dempster–Shafer theory based on functional levels and boundaries. In addition, the HFM is formed to indicate the level of problem functions and range area of the solution space. Components and parameters of harmful functions are obtained based on the digital twin entity. Creative ideas of product redesign are generated using the theory of inventive problem solving (TRIZ) to solve inventive problems at different functional levels. Technique for order preference by similarity to an ideal solution (TOPSIS) is introduced to evaluate and select solutions. Finally, the feasibility and effectiveness of the proposed method are verified in the redesign of an antenna mounted on vehicles.

数字孪生作为一种新兴的工业技术，在产品开发的各个阶段提供了巨大的机遇。在产品改进过程中广泛需要产品重新设计，这在很大程度上取决于产品的功能分析。传统的功能分析虽然可以识别产品设计问题，但分析的信息极其详细和冗长，阻碍了产品创新的机会。为了扩大解决方案空间以提高创新机会并确保产品在物理空间中的解决方案质量，在重新设计过程中引入了数字孪生。本研究提出了一种产品重新设计方法，该方法使用从关系函数模型 (RFM) 获得的功能回溯到具有数字孪生的分层函数模型 (HFM)。基于选定的目标产品，所提出的方法构建了源自反向鱼骨和组件之间关系的产品 RFM（子字段）。获取组件的相关参数。基于目标物理产品使用RFM（子字段）和参数构建数字孪生实体，并以“动词+名词”的形式提取功能。RFM 是考虑到函数之间的四种关系而形成的。此外，RFM 中的功能使用基于功能级别和边界的 Dempster-Shafer 理论划分为不同级别。此外，形成HFM以指示问题函数的级别和解空间的范围区域。基于数字孪生实体获取有害功能的组件和参数。产品重新设计的创意想法是使用发明问题解决理论（TRIZ）产生的，以解决不同功能层面的发明问题。引入了通过与理想解决方案相似的顺序偏好技术 (TOPSIS) 来评估和选择解决方案。最后，通过对车载天线的重新设计，验证了所提方法的可行性和有效性。