Scholarlink

<p>In the Industry 4.0 era, the rapid development of digital twin (DT) technology has produced a large body of machine-tool-oriented studies covering virtual commissioning, condition monitoring, process prediction, adaptive control and maintenance decision-making. This critical review first establishes a reproducible literature-screening procedure and then classifies machine-tool DT studies according to modelling strategy and lifecycle stage. Beyond summarising reported methods, the review distinguishes laboratory demonstrations, pilot-line validations and actual shop-floor applications, because the industrial maturity of MT4.0 remains uneven across factories. The analysis indicates that the transition towards MT4.0 remains incomplete; rather, many production sites are still moving from connectivity and data acquisition towards digital-factory-level integration. The main unresolved issues are cross-scale model coupling, real-time and transferable hybrid modelling, weak causal closure from state to quality and maintenance decisions, insufficient interpretability of black-box learning, and limited human-in-the-loop decision support. On this basis, MT5.0 is discussed as a goal-oriented perspective under Industry 5.0 rather than as a completed industrial stage. A knowledge-data-driven multidimensional twin framework and a deep human-machine collaboration framework are proposed, and practical limitations and future research priorities are identified, including deployment cost, real-time computation, knowledge acquisition, cross-platform interoperability and industrial validation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Digital twin of machine tools: a review and perspective

  • 期刊论文

  • 发表日期：

    2026年8月4日

查看全文
  • Wei Dai,
  • Guofeng Wang,
  • Xianlei Shan,
  • Decai Li,
  • Shilong Zhou

摘要

In the Industry 4.0 era, the rapid development of digital twin (DT) technology has produced a large body of machine-tool-oriented studies covering virtual commissioning, condition monitoring, process prediction, adaptive control and maintenance decision-making. This critical review first establishes a reproducible literature-screening procedure and then classifies machine-tool DT studies according to modelling strategy and lifecycle stage. Beyond summarising reported methods, the review distinguishes laboratory demonstrations, pilot-line validations and actual shop-floor applications, because the industrial maturity of MT4.0 remains uneven across factories. The analysis indicates that the transition towards MT4.0 remains incomplete; rather, many production sites are still moving from connectivity and data acquisition towards digital-factory-level integration. The main unresolved issues are cross-scale model coupling, real-time and transferable hybrid modelling, weak causal closure from state to quality and maintenance decisions, insufficient interpretability of black-box learning, and limited human-in-the-loop decision support. On this basis, MT5.0 is discussed as a goal-oriented perspective under Industry 5.0 rather than as a completed industrial stage. A knowledge-data-driven multidimensional twin framework and a deep human-machine collaboration framework are proposed, and practical limitations and future research priorities are identified, including deployment cost, real-time computation, knowledge acquisition, cross-platform interoperability and industrial validation.