Creating the ultimate onsite safety assistant using machine learning on construction projects

Creating the ultimate onsite safety assistant using machine learning on construction projects

By Lucy Pemberton

How has Skanska met their pledge to not only improve their construction safety, but to achieve a truly injury-free environment? The answer is through utilizing, a photo and video analytics platform that’s powered by machine learning algorithms. The platform analyzes all audio and visual data and automatically tags the data based on the content. This data is then, in turn, analyzed further using machine learning techniques, which detects and flags common safety risks such as missing gloves or hard hats.

This application of artificial intelligence (AI) to the construction sector has helped Skanska improve their construction safety, and if the rapidly expanding range of applications of AI to other sectors is anything to go by, the construction industry too will be following suit with a myriad of applications over the next 5 years.

Machine powered algorithms protecting workers:

Estimates suggest that three out of ten workplace injuries are a direct result of not wearing the correct Personal Protective Equipment (PPE), and these could be drastically reduced through proper automated monitoring and alerts. A survey conducted by the American Society of Safety Engineers (ASSE) in 2010 found that 98% of respondents had observed workers not wearing safety equipment when they should have been, and 30% had witnessed this on a number of occasions, showing that this is a wide-scale problem. The UK government’s innovation agency, Innovate UK, is providing $400,000 of funding into trials where footage at site entrances and exits are monitored by cameras connected to a real-time video analysis platform powered by machine learning algorithms.

As more construction developments are planned across the world, the innovative use of PPE monitoring could bring a real benefit to site safety, and in comparison to manual PPE checks, the technology could improve work-place productivity, whilst at the same time reducing the likelihood of human error.

Combining IoT with AI and machine learning to create a safer construction site:

Construction technology start-up, Pillar, has been pioneering AI-based solutions for the construction industry, and plans to aggregate data collected by wireless devices for the purpose of running predictive analysis algorithms. This innovation will also improve construction site safety through the benefit of collective insights provided by the numerous sensors deployed by Pillar in construction sites across the United States. For example, the aggregated data will be able to predict at what precise temperature a particular pipe will freeze on the first floor of a given building, compared with the third floor, and as the devices gather more information the predictive abilities will get better and better.

This is just the beginning

McKinsey has produced a thought piece entitled, ‘Artificial Intelligence: Construction Technology’s Next Frontier’, in which they compare the take up of AI technologies by industry. The article outlines that the applications of artificial intelligence are still in their nascent phases, and as such, AI isn’t used as extensively as it is in other sectors. As machine learning algorithms are beginning to be applied to construction industry use cases, it looks set that this will only increase over the coming years.

If you have any questions or would like to know if we can help your business with its innovation challenges, please contact us here or email us at

Never miss an insight

Get insights delivered right to your inbox

More of Our Insights & Work

Never miss an insight

Get insights delivered right to your inbox

You have successfully subscribed to our newsletter.

Too many subscribe attempts for this email address.