A case study by Suffolk Construction and image-analysis firm Smartvid.io into how machine-learning algorithms can improve safety and reduce risk has spurred the creation of the Predictive Analytics Strategic Council, an initiative by 10 major construction companies to share project data for A.I.-driven analysis. 

“This work is not about one company’s risk; it’s about reducing risk across our industry,” said Jit Kee Chin, chief data officer for Suffolk, in a statement announcing the formation of the group. In addition to Suffolk, other members include contractors Barton Malow, Clayco, DPR Construction, JE Dunn, Messer Construction, Mortenson, Shawmut Design and Construction and Bouygues Group. Risk and insurance broker Aon is also on board, and Smartvid.io will serve as technical advisor. The firms have agreed to share digital project data from a period of more than 10 years to build predictive models for safety and risk. The council is not financially backed by any vendors, although some of the participating firms will provide some funding.

“The first order of business is to build on the work started with Suffolk where we built these predictive models of safety incidents,” says Josh Kanner, founder and CEO of Smartvid.io. “The second part is to understand the operational impact and implications of these predictions.” Results will be public, and the council is accepting new member applications.

Since firms are reluctant to reveal incident reports and site photos that may not reflect well on their practices, the data is anonymized before being fed into the analytics. But the potential for new insights is worth the effort, Kanner says. “We’ll have improved models as we move into this world where machine learning and operational intelligence work in production.”