Session Aims & Scope
Digital twins are revolutionizing the urban built environment by creating dynamic, digital replicas of physical spaces. These virtual models integrate data with different sources and formats, enabling city planners and managers to simulate, monitor, and optimise urban systems with unprecedented precision. The importance of digital twins lies in their ability to enhance operational efficiency, predict maintenance demands, and support sustainable urban development. Digital twins in the urban built environment is promising in deeper integrations with AI for predictive analytics, expanded use in fields like energy conservation and disaster management, and broader adoption in smart city initiatives to improve the quality of life for urban residents.
The session will focus on the following points
- Digital twins in urban planning and development
- Digital twins in property and facility management
- Digital twins in architecture and built environment design
- Digital twins in sustainable energy management
- Digital twins in disaster management
- Smart city initiatives
Session Chair(s)
Chair
Cecilia Bolognesi
Professor, Politecnico di Milano (Italy)
cecilia.bolognesi@polimi.it
Co-Chair
Qian-Cheng Wang
PhD Candidate, University of Cambridge (UK)
qw250@cam.ac.uk
Co-Chair
Xuan Liu
Doctoral Candidate, Eindhoven University of Technology (Netherlands)
x.liu1@tue.nl
Session Presentation
1.
Bige Tunçer
Professor
Eindhoven University of Technology (Netherlands)
Title: Information Systems in the Built Environment: Driving Climate Action and Meaningful Transformation through AI and Data-Driven Technologies
Abstract
This talk explores how information systems in the built environment can drive climate action and transformative change through AI and data-driven technologies. It highlights the role of multi-source, multi-scale, and multi-time data on built environment parameters, user patterns and opinions, and sensible environmental characteristics can yield actionable insights and socio-technical solutions for big issues. The talk will present methodologies for creating intelligent design support tools, showcase case studies, and discuss the potential of data-driven strategies and AI to foster climate resilience, enhance user experience, and support decision-making for a more sustainable built environment.
2.
Cecilia Bolognesi
Professor
Politecnico di Milano (Italy)
Title: Research for the construction of digital twins for the knowledge and maintenance of historical architectural heritage
Abstract
The issue of the maintenance of Italy’s valuable historical built heritage often must reckon with the size of publicly owned assets or those of religious bodies. In cases such as these, the construction of management models useful for maintenance with efficient processes passes obligatorily through the digitisation of the heritage first and the construction of automatisms proper to the digital twin. In this presentation, we will touch upon three nodal points of this theme: the use of a collaborative environment for multi-user sharing of synchronically annotated information, monitoring of the influx of people with feedback in the digital model, structural monitoring and automatic quantification of materials and works for maintenance.
3.
Li WAN
Associate Professor
University of Cambridge (UK)
Title: From urban modelling to city digital twins – Some reflections
Abstract
The fast advancement of digital twins (DTs) in engineering and manufacturing is now driving an economy-wide technological revolution. The success of DT applications in manufacturing demonstrates the power of data and machine learning techniques in unlocking significant performance improvements and efficiency gains. Upscaling engineering DTs to city digital twins (CDTs), however, faces several inherent challenges. DTs are most useful when the problem and solution space resemble a well-definable, closed engineering system. Unfortunately, most societal challenges do not fit this description. Existing big data sources for cities, despite unprecedented in terms of their volume and velocity, may still be ‘tiny’ relative to the sheer complexity of urban systems and processes. The effectiveness of a purely data-driven approach in CDT development might thus be overstated. Given the substantial knowledge gap in understanding how cities function and rising concerns about data privacy and AI ethics, it is thus questionable to assume that CDTs can and should pursue the same level of analytical and automation capability seen in engineering and manufacturing sectors. Enhancing a positive feedback loop between data and CDT models, particularly by informing data collection through theory-driven model calibration and validation, seems a promising way to advance CDT models.
4.
Wei-Ting Hong
Research Associate
University of Sydney (Australia)
Title: Automated analysis of infrastructure policies using natural language processing (NLP)
Abstract
This study presents an NLP-driven framework for analysing infrastructure policy trends over time to support policymaking. By leveraging NLP techniques, the framework identifies and classifies key infrastructure policies, offering strategic insights and flexibility for further statistical analysis. The proposed infrastructure policy strategic map categorises policies based on trends and frequency of mentions, providing an efficient tool for policymakers. A case study of New South Wales (NSW) infrastructure policymaking highlights shifts in policy priorities. This scalable framework can be applied to other jurisdictions, enabling efficient analysis of large volumes of infrastructure policy data.
5.
Yuxin Zhang
Research Assistant Professor
The Hong Kong Polytechnic University (China)
Title: Smart Evacuation Systems Using digital twin and AI in Building Fires
Abstract
The development of real-time smart evacuation systems is vital for enhancing public safety during building fires, providing a crucial societal benefit. This study introduces an intelligent Emergency Digital Twin (EDiT) system enabling real-time tracking, speed estimation, and privacy protection of evacuees. Achieving detection success rates of 100% for individuals and 93% for multiple objects, the system significantly improves evacuation efficiency. Moreover, the creation of the Human Behaviour Detection Dataset (HBSet) enhances the ability to identify vulnerable groups, such as the elderly and disabled, ensuring they receive necessary assistance during emergencies. Additionally, a novel multi-camera tracking system is introduced, offering seamless monitoring across complex buildings. The integration of AI and digital twin technology in emergency management not only optimizes evacuation strategies but also contributes to a safer, more resilient urban environment. This advancement aids authorities in making informed, timely decisions, ultimately reducing casualties and property damage in fire scenarios, representing a critical societal safeguard.
6.
Cuicheng Zhang
PhD Researcher
University of Oxford (UK)
Title: Data-Driven Energy Performance Analysis: Insights from London’s Residential Buildings
Abstract
This research leverages digital twin technology to examine the relationship between socio-demographic factors and building attributes on energy efficiency and environmental impacts in London’s residential buildings from 2011 to 2021. By utilizing Energy Performance Certificates (EPCs) and socio-demographic data, it integrates complexity science and data-driven methodologies, including machine learning models like GLM and XGBoost. The study reveals that factors such as income, employment, and cultural diversity significantly affect energy efficiency. Through the digital twin framework, the research suggests digital twin oriented adaptive energy policies that account for socio-demographic changes, aiming to guide sustainable urban development and enhance environmental outcomes.
7.
Qian XU
Lecturer (Assistant Professor)
University of Lincoln (UK)
Title: Integrating Digital Twin Technology for Sustainable Urban Development: A Focus on Transit-Oriented Development
Abstract
The rise of Digital Twin technology offers significant potential for enhancing sustainability in urban development, particularly in the context of Transit-Oriented Development (TOD) and compact city frameworks. This presentation introduces a novel approach that integrates Digital Twin systems into the Node-Place-Ecology (NPE) model to simulate, monitor, and optimize TOD sustainability across urban environments. By creating real-time digital replicas of TOD station areas, the model facilitates a dynamic and accurate analysis of transportation, land use, and ecological factors.
Utilizing the Tokyo Metropolitan Area as a case study, this research demonstrates how Digital Twin technology enables continuous monitoring and real-time updates of urban development, enhancing the ability to address challenges such as shifting population densities, economic changes, and environmental impacts. The integration of Digital Twin allows for precise tracking of TOD performance, helping urban planners and policymakers simulate future scenarios and develop data-driven strategies to improve sustainability outcomes.
This presentation will explore how the combination of Digital Twin and the NPE model creates a comprehensive framework for managing the complexities of modern cities, ensuring long-term resilience and adaptability in the face of rapid urbanization and environmental change.