Digital Engineering, Data and Computer Vision - Past projects

Project:  Towards Data-centric Smart Infrastructures as Cyber-physical Systems

The first project aims to identify the information need, model the data and prototype a digital twin of Staffordshire Bridges in collaboration with Microsoft and Network Rail. Didem's research further extends to develop a framework which aims to provide an information platform for digital twins and cyber-physical systems, including bridges, railways, roads, buildings and other similar assets. Didem uses systems thinking, design thinking and futures thinking to support stakeholders as they seek to identify the key data resources and information needed to better integrate different digital twins, where interoperable and sustainable decisions could be possible. For this purpose, Didem conducts interviews with partner organisations to support the framework.

Additionally, Didem leads a study that is focused on planning future scenarios for the built environment of the UK in 2040 - together with a team of CDBB postdoctoral researchers. They work together to identify relevant trends and uncertainties (from people, mobility, infrastructure, technology, economy, government, and resources dimensions) and evaluate the intersection of these factors to build different scenarios. This study does not only use scenarios to initiate discussions on how the industry should act in rapidly changing, post-COVID19 world but also suggests strategies for enabling more sustainable future built environment.

Sponsor: CDBB - CIH

Partners: Cambridge Centre for Smart Infrastructure and Construction (CSIC)

Laing O’Rourke Centre involvement: Research Associate: Dr Didem Gurdur Broo, Academic Supervisor Prof. Cam Middleton

Dates: November 2019 - September 2021

Online material for further information:

• Popular science article: Tech firms are winning the AI race because they understand data – other sectors need to catch up

• Popular science article: Let Covid crisis be catalyst to change construction for good

Project: Cloud-based Building Information Modelling

Building Information Modelling (BIM) as a product and process enables stakeholders across the built environment sector to create digital versions of real world assets (such as buildings, bridges and tunnels). The digital versions are commonly called 'digital twins'. When placed on the cloud, the digital twins can serve as a resilient and integrated repository of all asset data throughout their life-cycle. Such a repository is a key enabler in this sector of all upcoming IT waves, such as cloud computing, data analytics, participatory sensing, and smart infrastructure. The potential benefits have attracted interest from a wide array of end-users whose interests span from early design phases to operation and asset management, and from roads and bridges to industrial off-shore facilities. This has led to aggressive market penetration in the last decade. However, the full potential of BIM is currently exploited only in a fairly narrow range of applications. This is mainly due to the lack of trained scientific personnel capable of understanding the value of BIM and creating the link between digital twins and possible applications.

The ambition of CBIM is therefore to educate researchers in the development of a set of novel and disruptive BIM technologies that will automate the generation and enrichment of digital twins, improve the management, security and resilience of BIM-enabled processes, and boost the industrial uptake of BIM across sectors and disciplines by training these researchers to valorise and exploit their work. This new generation of researchers can play a key role in the widespread implementation of BIM products and processes dedicated to digitising our built infrastructure and managing our assets better to yield massive gains in sustainability, productivity and safety. [Further information]

Sponsor: European Union's Horizon 2020 research and innovation programme - H2020-EU.1.3.1.

Partners: Technion Israel Institute of Technology, University of Cambridge, Technische University Berlin, University College Dublin, University College London, Trimble Oy, Fundacion CARTIF, LocLab Consulting GmbH

Laing O’Rourke Centre involvement: Deputy Coordinator Dr Ioannis Brilakis

Dates: March 2020 – February 2024

Online material for further information:

• Get involved: Building Information Modelling School - online, from 7-11 December 2020

Project: Application of satellite technology in infrastructure monitoring

The satellite monitoring research group has been established in the Laing O'Rourke Centre by Sakthy Selvakumaran to build upon research into remote sensing for urban environments, specifically the use of satellite imagery for infrastructure monitoring. [Further information]  

Sponsors and Partners: Sakthy was awarded the Isaac Newton Trust/Newnham College Research Fellowship in Engineering, which she is using to establish and lead a growing research group in this field. There is considerable interest from government and industry partners in this project. Plans for this emerging group are developing with various collaborators, including The Engineering and Physical Sciences Research Council (EPSRC), The Centre for Digital Built Britain (CDBB) and Centre for Smart Infrastructure and Construction (CSIC).

Laing O’Rourke Centre involvement: Academic Supervisor Prof. Cam Middleton, Principal Investigator Sakthy Selvakumaran, Senior Research Associate Dr Gabriel Martin

Dates: October 2019 – ongoing

Project: Artificial Intelligence Optimised Pathways for Schedule Execution

The project seeks to develop a novel automated 'schedule learning platform' that applies data science and machine learning to thousands of previous project schedules, offering a unique scalable solution for improved certainty and confidence in project planning for future projects. The solution is based on thousands of previous construction projects, allowing the platform to learn across projects what was planned to happen and what actually happened, thus reducing the effect of human bias, subjectivity and inaccuracy. Schedule data is analysed, similar tasks and relationships are automatically grouped, with patterns drawn using Artificial Intelligence, enabling the platform to predict the most likely outcome for every task and provide optimal paths/recommendations to mitigate risks/delays. [Further information] 

Sponsor: Innovate UK

Partners: nPlan, Kier

Laing O’Rourke Centre involvement: Investigator Dr Ioannis Brilakis, Research Associates Sally Xie and Ying Hong

Dates: March 2019 - February 2021

Project: Interactive Point Cloud and Image Data Generation in Infrastructure Scenes

All construction trades and inspectors need spatial and visual data to operate. However, extracting up to thousands of measurements and creating meaningful registered data sets in a single work shift translates to a sizable portion of work time dedicated to measuring and processing instead of direct work. The research is mainly focused on increasing the efficiency of geometry and visual data collection by devising a system that tackles the challenges in data collection and post-processing through a combination of high-resolution DSLR cameras and a hand held laser scanner.

Construction operatives and inspectors could benefit significantly from laser-scanner-quality data in terms of resolution and accuracy, with the mobility and reachability of hand held devices and the ability to operate outdoors reliably and from long distances.

Sponsors and Partners: BP, Laing O’Rourke, Topcon, GeoSLAM, Trimble

Laing O’Rourke Centre involvement: Principal Investigator Dr Ioannis Brilakis, PhD student Maciej Trzeciak

Dates: October 2017 – September 2021

PhD Project: Automated BIM Generation of Existing and Under Construction Railways

This research is aimed at creating a viable approach to automate the generation of as-is geometric railway Building Information Modelling (BIM).The key novel idea that makes this possible is that railways follow a set of engineering design assumptions which can be used as guides for segmentation by Markov Chain Monte Carlo (MCMC) methods, region growing, or octrees.

Researcher: Mahendrini Fernando Ariyachandra

Supervisor: Dr Ioannis Brilakis

Funding: Bentley Systems UK Ltd, Cambridge Trust

Dates: October 2017 – September 2021

Online material for further information:

• Update: Automating the process of creating digital twins of rail infrastructure

Project: Automatic Construction Monitoring Through Semantic Information Modelling

This project aims to develop computational algorithms and methods for automatic as-built construction monitoring through semantics-based Building Information Modelling (BIM). Construction as–built monitoring is crucial for the cost, time, quality and safety of projects. Methods for generating as-built status are primarily manual. There are gaps in sophistication of automation, and recognition for semantic construction information during the process is low. The project is expected to provide efficient and accurate solutions for as-built construction monitoring. The implementation of the as-build modelling approach will provide a significant tool to manage the status of construction project, including progress and quality tracking. [Further information]

Sponsor: Australian Research Council

Partners: Curtin University, Australia

Laing O’Rourke Centre involvement: Co-Principal Investigator Dr Ioannis Brilakis

Dates: April 2017 – March 2020

PhD Project: Top-down As-is Modelling of Industrial Facilities

This research will explore ways to detect existing building objects using spatial and visual data for the purpose of automating the generation of as-built geometric models of industrial facilities. The research will address both construction engineering and computer vision issues.

Researcher: Evangelia Agapaki

Supervisor: Dr Ioannis Brilakis

Funding: EPSRC (DTP) and AVEVA

Dates: October 2016 – September 2020