Title: Twin Systems
Prof. Ioannis Brilakis
University of Cambridge, UK
Abstract
Digital Twinning methods can produce a reliable digital record of the built environment and enable owners to reliably protect, monitor and maintain the condition of their asset. The built environment is comprised of large assets that need significant resource investments to design, construct, maintain and operate them. Improving productivity, i.e., efficiency and effectiveness, and creating new, disruptive ways to address existing problems throughout their lifecycle can generate significant performance improvements in cost, time, quality, safety, sustainability, and resilience metrics for all involved parties. Creating and maintaining an up-to-date electronic record of built environment assets in the form of rich Digital Twins can help generate such improvements. This talk introduces research conducted at the University of Cambridge on inexpensive AI methods for generating object-oriented infrastructure geometry, detecting, and mapping visible defects on the resulting Digital Twin, automatically extracting defect spatial measurements, and sensor and sensor data modelling. The results of these methods are further exploited through their application in design for manufacturing and assembly (DfMA), mixed-reality-enabled mobile inspection, and proactive asset protection from accidental damage.
Prof. Dongping Fang
Tsinghua University, China (President, CIB)
Title: Human-Centric Approach for Urban Resilience
Abstract
Cities are faced with significant threats from natural hazards, necessitating the enhancement of urban resilience. However, as urban systems grow increasingly complex, these hazards not only cause casualties and direct damage to physical elements such as buildings and infrastructures but also trigger cascading failures within engineering systems and extend impacts to social dimensions like healthcare and human mobility. This talk proposes human-centric and data-driven approaches to examining urban systems and their resilience issues. By conceptualizing a city as 'a system of systems under trio-spaces', a cross-system and cross-dimension approach based on scenario deduction is proposed for analysing urban resilience. Computing models are employed to capture the dynamics of residents’ needs as well as urban functionality in the aftermath of natural hazards, which contributes to determining key infrastructure to be protected and optimally allocating the limited resources to various systems in the city. By leveraging a hypernetwork model, hyperedges are introduced to capture the interactions among social systems, infrastructure, and human needs. This enables a comprehensive understanding of post-disaster recovery processes for people-centric approaches. The latest work in modelling spatial behaviours in response to extreme events will also be presented. By integrating spatial machine learning methods with interpretable analytics, a Geographically Weighted Random Forest model is proposed to characterize localized spatial relationships. Case studies will be provided to demonstrate the feasibility and effectiveness of the proposed approaches for analysing urban resilience.
Title: From Bytes to Benefits: Digital Transformation for Managing Risk in Construction
Prof. Roger Flanagan
University of Reading, UK
Abstract
The construction industry has become more complex, with bigger projects, innovative and exciting designs pushing the boundaries of technology, global turbulence leading to disruption in supply chains, and increasing environmental, social, legal, regulatory, and compliance pressures. New risks are emerging and converging, such as cybersecurity, climate change, and integrating modular integrated construction (MiC) on site, which requires new approaches to risk management. Risks are no longer isolated, their convergence is creating complex, systemic challenges that demand integrated, cross-functional responses. Risk management, rather than eliminating uncertainty, aims to reduce the negative effects of risks. With more uncertainty there is the need for better risk management processes to cope with the increased complexity. The traditional approaches of risk identification, risk analysis, and risk allocation producing a risk register are outdated in such a fast-changing and complex industry. Unlike manufacturing with its controlled environments, construction projects are complex ecosystems with countless variables from weather conditions to regulatory requirements, to multiple stakeholders with competing priorities. Addressing the speed and complexity of the current risk convergence requires a new mindset and shifting attitudes to risk. Many enterprises are still in the early stages of adopting data and technology systems at point solution (single point of data) and single project levels. Digital tools and AI are beginning to help with the use of sensors, drones, and predictive analytics but there is insufficient integration. Big data and AI are the most important requirements for projects by using stress test decisions to manage risk. There is the need for re-alignment and integration across systems, with digital and AI at the core to improve risk management. Convergence across design and site production process is happening, but the construction industry has been slow to recognise the value of data, how it is collected, analysed, and managed; it has been inherently inefficient in the collection and analysis of data. AI and data have the potential to be the most important part of any enterprise. The insurance sector have been alerted to the increased risks they are being required to provide as liability claims increase in size. There needs to be an awakening to how data holds the key to better risk management. Changing procurement systems demand greater integration by engagement with all stakeholders, including the supply chain, moving away from the silo mentality. AI and digital tools have helped improve efficiency; the next step is to rethink how risk management can be better integrated. Risk must be realigned as a value driver - embedding data-driven insights into decision-making to build long-term resilience. Keyword: risk management, AI and data, convergence, complexity, integration.
Title: Leveraging Multimodal Large Language Model (MLLM) to Enhance Object Detection, Communication and Collaboration of Mobile Robots
Prof. Kincho Law
Stanford University
Abstract
Large language models (LLMs) are powerful resources that can effectively be applied to perform many useful tasks. Mobile robots equipped with cameras, lidar and other technologies are widely available and deployed to navigate and interact with their environments in real-time. This presentation discusses a prototype mobile robot platform to test the utilization of multimodal LLMs for object detection, human-robot and robot-robot communication and collaboration.
Title: Ai-Empowered Sustainable Construction
Prof. Geoffrey Shen
The Hong Kong Polytechnic University
Title: Humanization in an AI-Driven Construction Industry: Hora Est…?
Prof. Wilco Tijhuis
University of Twente, Netherlands
Abstract
In nowadays construction industry the developments towards an Artificial-Intelligence (AI)-driven business-model are strongly supported by the several technological innovations in hardware and software. Think e.g. about intelligent building information modelling (e.g. BIM with more than 4- or 5-D-models), data-integrated technologies to support energy-savings of the built environment, security and safety in buildings and on construction-sites, robotized (pre)fabrication factories for complete housing projects, etc. All these developments are increasingly signalling that the construction industry of the future is not lead by construction professionals and their businesses, but far more by IT(application)-professionals and their businesses, e.g. originating from several kind of ‘silicon-valley’ regions of the world. Challenging times and with a lot of new opportunities, for sure! However, where is the human being in all these developments? Do we really know our workers at the site and in the factories? Are they still professionally skilled, or simply equipped to become a kind of ‘extension of robotic arms’, i.e.: How to deal with so called ‘last mile logistics’ in these? And, not in the least, what about the client? Does construction industry really deliver ‘more value for money’, now technology obviously makes it quite easily possible, but also causing a huge demand of complex IT-investments? All serious considerations, where we therefore need to put the question forward: Is there still a serious position in this AI-Driven Construction Industry for a kind of ‘humanization’, or has the time for this meanwhile passed? Just like the famous old classic Roman saying: Hora Est…?
Title: Ways That Ai Innovation in Smart Cities™ Construction Can Improve Infrastructure Resilience
Prof. Liz Varga
University College London
Abstract
Innovation in the use of Artificial Intelligence (AI) methods in the built environment have burgeoned in recent years. They build on the growth of smart cities and the related analysis of vast data streamed from sensors, GPS devices, smartcards, CCTV, mobile phones, and social media that have enabled the development of digital twins and simulation models. The use of AI is improving predictions from this vast data and providing insights for many smart city construction projects including workforce safety; construction equipment optimisation, and project task optimisation. Yet the extent to which AI is innovated to improve infrastructure resilience (the ability to absorb and adapt, to hazards and threats) during construction in smart cities is much more limited and provides a significant opportunity to embrace infrastructure resilience within the smart city concept.
Title: Research of Digitalization in Construction: Where Are We?
Prof. Shuibo Zhang
Tianjin University
Abstract
I will, based on an integrated review of key journals in construction and project management, answer the following 3 questions: 1. What is the evolution of the thematic research trends in digtial construction? 2. what are the research gaps and future research directions? 3. What are the policy implications?