AI scanning a building web site can detect when issues are falling behind
When managers visit a site once or twice a week, the camera on their head captures video footage of the entire project and uploads it to image recognition software that compares the status of many thousands of objects on site, such as: B. Sockets and bathroom fittings – with a digital replica of the building.
The AI also uses the video feed to track within inches of where the camera is in the building so it can identify the exact location of the objects in each image. The system can track the status of around 150,000 objects several times a week, says Danon. For each object, the AI can recognize which of three or four states it is in, from not yet started to complete installation.
On-site inspections are slow and lengthy, says Buildots' Sophie Morris, a civil engineer who worked in construction before joining the company. Buildots AI eliminates a lot of repetitive tasks and allows people to focus on making important decisions. "This is the job people want to do – they don't have to look to see if the walls have been painted or if someone has drilled too many holes in the ceiling," she says.
Another plus is the way the technology works in the background. "It collects data without having to search the website with spreadsheets or schedules," said Glen Roberts, operations director at Wates. He says his company is now planning to roll out the Buildots system in other locations.
Comparing the full status of a project to its digital plan several times a week also made a big difference during the Covid-19 pandemic. When construction sites closed to everyone but key local staff, managers of several Buildots projects were able to remotely track progress.
But AI is not going to replace these important people anytime soon. Buildings are still made by people. "At the end of the day, this is a very work-oriented industry and that won't change," says Morris.
Change note: We changed the text to show how the Buildots system differs from others.