MAKING CONSTRUCTION PROJECTS SMARTER WITH A DIGITAL TWIN

Digital twins and AI help to bridge the construction productivity gap  

Bam Nuttall, Cranfield University and Iotic create The Learning Camera

In a drive to increase on-site productivity and operational performance in the construction industry, BAM Nuttall has teamed up with tech-firm Iotic and researchers at Cranfield University to develop an AI-based, computer-vision system using new digital twin technology.

The Learning Camera project employs a standard webcam, integrated with an IoT framework of smart sensors to collect real-time environmental data such as wind speed and weather conditions, combined with contextual information including location, date and time. All this data is fed into a cloud-based system to create digital twins, which bridge the physical and virtual world.

An Iotic digital twin is an autonomous and interoperable version of a ‘thing’ or an asset with all its data and controls that can interact, interrelate and behave in a digital environment as its twinned counterpart in the real world. Pairing these two worlds with the Learning Camera enables the monitoring and analysis of on-site data to identify and head off problems before they cause potential delays, accidents and increased costs, while also helping with future planning by being able to run accurate simulations with real data.

In particular, cameras can be programmed to recognise abnormal activity on a construction site 24 hours a day and generate alerts so that someone can attend to rectify any problems. This reduces the need for repetitive on-site checks and security monitoring in hazardous areas and all-weather conditions.

Sophie Peachey, Head of Customer Success at Iotic, said: “The application of Digital Twin technology within the Learning Camera allows us to broker access to a potentially increasing number of data sources and controls to perfect the accuracy of the algorithms used in the solution. These algorithms must be able to interpret differences correctly and instigate appropriate actions to make the Learning Camera a solution that people trust. You can see how this could apply to different situations in which people have to balance the importance of knowing that something is there, has changed, or is working, against the cost of their time in checking. While this is not restricted to construction, we are very excited by the impact this could have on productivity and in providing construction staff with a safe working environment.”

Dr Yifan Zhao, lecturer in Image and Signal Processing and Degradation Assessment at the Through Life Engineering Service Institute at Cranfield University, said he believed the innovation was a great opportunity for AI to be applied to a traditional industry. “By using The Learning Camera, construction sites will be better equipped to manage and deliver projects. Its use will also promote the need for the industry to attract talent with skills in software and hardware development in order to tackle the much-publicised poor productivity levels.”

And Colin Evison, head of innovation at BAM Nuttall, concluded: “Overall, lessons are learned and opportunities are uncovered within the virtual environment that can be applied to the physical world and used to transform a business. This is a real opportunity to explore how we can make construction projects smarter by the adoption and development of technology solutions which have not been traditionally available before.”