APSIRI Research: From Digital Twin to Digital Cognition in Smart Manufacturing
Smart manufacturing has advanced rapidly on the strength of the “digital twin” — a virtual model that mirrors a physical system in real time. But mirroring, however sophisticated, is not the same as understanding. A new research outcome from the Asia-Pacific Social Innovation Research Institute (APSIRI), led by Xiaole Ruan together with Yiyang Zeng, Hongjian Shao, Zhipeng Cai, Ying Ying, Junjiang Jin, Jijun Lin, and Enwei Ni, proposes a multilevel model of algorithmic intelligence that charts a path from digital twin toward “digital cognition.”
The distinction at the heart of the study is conceptually important. A digital twin represents what is happening in a physical system; it reflects the present state of a machine, a process, or a factory. Digital cognition goes further: it implies the capacity to interpret, reason about, and act intelligently upon that information — to move from representation toward something closer to understanding. The research conceptualises how manufacturing systems might make this transition, advancing through successive levels of algorithmic intelligence.
The multilevel structure of the model is one of its key contributions. Rather than treating intelligence as a single threshold a system either crosses or does not, the study describes a layered progression — from systems that faithfully mirror physical processes, through systems that analyse and predict, toward systems capable of higher-order, cognitive functions. This layered view offers a clearer map of where current technology stands and where it might develop, and a vocabulary for discussing capabilities that are too often lumped together under the single word “smart.”
The implications for industry are substantial. As manufacturing systems acquire more sophisticated intelligence, they promise gains in efficiency, adaptability, and resilience — the ability not merely to monitor operations but to anticipate problems, optimise processes, and respond intelligently to change. Understanding the path from digital twin to digital cognition helps clarify what such systems can realistically achieve at each stage, and what advances are required to move forward.
The research also raises questions that reach beyond engineering. As algorithmic systems take on more cognitive functions, questions arise about the appropriate relationship between human and machine intelligence — about which judgements should remain human, how oversight should be maintained, and how the benefits of advanced automation can be realised responsibly. A clear conceptual model of machine cognition is a necessary foundation for thinking carefully about these questions.
By offering a rigorous framework for understanding algorithmic intelligence in manufacturing, the study contributes to both the theory and the practice of digital transformation. It helps move the conversation beyond loose talk of “smart factories” toward a more precise understanding of what machine intelligence is, how it develops, and what it makes possible.
This research contributes to APSIRI’s work on digital transformation and inclusive technology. By examining how intelligent systems develop and what their advance means, the Institute seeks to support a digital economy in which technological progress serves broad and sustainable human benefit.