Beyond connected digital twins – Can digital twins really deliver sustainable cities?

We examine the evolution of digital twins and introduce the ideas underlying The World Avatar (TWA), moving from a representation of physical entities to an approach founded on the idea of a general knowledge representation of the world at large. The ‘world’ in TWA refers not to planet Earth but to the space of all possible concepts. Unlike traditional digital twins, which often struggle with interoperability due to siloed data and proprietary standards, TWA is designed to enable the representation of any concept, allowing it to embrace any domain in terms of things and actions. This is achieved by leveraging semantic technologies, including ontologies, knowledge graphs and semantic agents in combination with machine learning to address the issue of interoperability. We show examples relating to flood risk modelling, energy planning and social equality assessments. The examples show how the design of TWA has been used to achieve scalable and seamless integration of data, models and computational agents to overcome the limitations of siloed approaches, and how the digital twinning approach underlying TWA can contribute to addressing urban development challenges to provide cross-domain insights that allow actionable, data-driven decision-making. The ideas underlying TWA are proposed as the ultimate evolution of digital twinning approaches, laying the foundation for scalable, interoperable solutions aligned with global sustainability goals, with particular relevance for sustainable urban development. • Examine the evolution of digital twinning approaches. • Argue the benefits of explicit knowledge models to enable semantic interoperability. • Explain The World Avatar (TWA) in the context of connected digital twinning. • Show examples of TWA overcoming data silos across social and technical domains. • Demonstrate potential to inform planning and sustainable development.