Computer Vision AI revolutionises safety and efficiency

Our CEO, Salman Chaudhary talks to Construction Computing about why those companies that move early with regards to Computer Vision (CV) AI will set the pace of the industry and gain prime benefits.

“Around 7 per cent of the world’s workforce is employed in construction activities and whilst other sectors have embraced AI, until recently at least, sites have lagged behind. This is against the background of a report by McKinsey which highlighted that total economic output per construction worker has remained flat, compared to productivity gains of 1,500 percent in manufacturing and agriculture since 1945. It points out that one of the reasons for this is that construction is still an under-digitized industry and slow to adopt new technologies (McKinsey).

However, all that is changing with Computer Vision AI having the ability to make construction sites more efficient and safer, whilst saving costs. AI offers a transformative change from a laissez-faire approach that relies on construction workers following general health & safety rules to monitoring, detecting and reporting breaches on COVID social distancing and face mask wearing.

When COVID-19 hit, the construction sector faced significant challenges in terms of securing supply chains and dealing with uncertain customer demand. Workplace safety suddenly became an incredibly significant challenge for a lot of construction businesses. Most organizations implemented manual monitoring of social distancing and wearing of face masks as standard operating procedures, although the level of compliance has been patchy. Without active and comprehensive monitoring of how the guidelines are being followed, it’s virtually impossible to know whether the risk of infections was being reduced.

As a consequence, very early in the pandemic there were some large outbreaks on site that lead to shutdowns and inevitable delays to the build schedule. At the same time, we had been developing Computer Vision AI that focuses on visual data and algorithms that can be applied to images and videos. The goal was to help site management teams to accurately and comprehensively monitor how well workers were complying with the safety guidelines they had put in place. The technology does this by applying AI models on CCTV camera video feeds to detect proper face mask usage, social distancing, and crowding within a workplace. This data is then analysed and assessed to provide specific insights into where improvements to workplace safety can be made.

Capturing this data – a task that would be impossible using observation and spot checks by Health & Safety Officers – provides a detailed assessment of the risk of infection, PPE compliance and potential disruption to a business. This reassures businesses by reducing the risk of infection and it also gives employees confidence that they are working in a safe environment, protecting themselves and their families.

As we move beyond COVID-19, Computer Vision AI will remain a disruptive technology in construction. The possibilities extend to risk mitigation (time, cost and safety risks), project planning, through to monitoring that workers are wearing correct PPE, for example, a safety harness. The opportunity to leverage AI and Computer Vision is phenomenal. It has the ability to improve most aspects of site operations, whilst reducing cost.

Case study:

We recently deployed our Computer Vision AI WorkSafe solution in a large plant. The system leverages existing surveillance and security assets. KPIs were set to monitor staff mask usage improvement and social distancing requirements around COVID-19. The aim was to provide rapid and targeted insights into where the business could improve its already robust occupational health policies to reduce risk to its employees and avoid operational interruptions.

Following the 6-week pilot project, which began in November 2020, the business recorded a multi-fold improvement in monitoring (spot checks vs Computer Vision) and witnessed more than 80% reduction in social distancing violations within first 4 weeks of pilot deployment and up to 90% reduction in mask noncompliance towards the end of the pilot.


There are huge opportunities for Computer Vision AI, especially for activities that require human inspection and are repetitive. It has the ability to make businesses more productive, to reduce their operational costs and carbon footprints, and to keep employees safe and businesses protected.

Despite concerns about job losses, AI is unlikely to replace the human workforce. Instead, it will alter business models, avoid expensive errors, reduce worksite injuries, and make construction sites more efficient.

However, construction is still behind the curve when it comes to implementing AI solutions, although it is catching up rapidly, mainly as a result of the pandemic where digitisation helped reduce the risk of infections and enabled sites to become fully operational quicker. Unlike other ‘point’ solutions, which offer limited coverage at entry points or procedural and manual checking, Computer Vision AI platforms provide insights on how effective current measures are, and, importantly, where improvements can be made over time.

It also enables employers to see where they need to do more with education, awareness or physical measures to improve safety. We believe this combination of real-time monitoring and measurement of effectiveness will bring confidence in a future of work which is safer for all. As a business, our goal is to help construction lead the digital revolution, moving it away from basic IT tools to AI and computer vision.”

Find out more about WorkSafe Analytics 

Leave a Reply

Your email address will not be published.