Oracle announced Dec. 16 it has completed acquisition of Newmetrix, a cloud-based platform that uses artificial intelligence to highlight construction problems focused on project safety.
Terms of the deal were not disclosed. Newmetrix founder and CEO Josh Kanner will become an Oracle executive and much of the team that created the platform will move to Oracle.
"This was an opportunity for us to ingest some pretty important technology and consume some intellectual property that really extends and expands our Construction Intelligence Cloud," said Mark Webster, Oracle senior vice president and general manager, construction and engineering. "This presents an opportunity immediately for us to accelerate our capabilities in using AI and machine learning to create more predictive behaviors."
Founded as Smartvid.io in 2015, Newmetrix has been integrated with Oracle's Aconex project management platform since 2019. Industry customers such as Suffolk Construction use its AI, called Vinnie, to identify and analyze risk based on construction site images captured during project documentation. Vinnie is programmed to detect 100 different safety risks from construction images. Indeed, Suffolk Construction was an alpha tester and user of the platform since its inception as Smartvid.io.
Kanner described the journey Suffolk and Newmetrix took as one applicable to the entire construction industry in understanding how josbite data can become more valuable. Kanner and the Newmetrix team trained its AI, Vinnie, to detect things like people, work boots, personal protective equipment, hard hats, tools and site equipment.
"How could it be used?" he said. "Where we started was in having photo and video management as a core part of the platform. Over time, what we saw is it was how the AI could unlock insights from that visual data."
Kanner said the data insights expanded over time and the company changed its name to Newmetrix precisely because of its ability to track such site metrics.
"In addition to looking at just photo data, we also started looking at structured data, too," he said. "How many safety observations were being done and what was the ratio of man-hours worked? Of supervisory oversight to apprentice oversight or trade partner hours, schedule activities. And that was with Suffolk really at the center of that."
The safety outcomes Newmetrix was trying to predict, in the data science world, are known as target variables, Kanner said. To accurately understand those variables, Suffolk provided 10 years of outcome data from its sites. Kanner also said that Jit-Kee Chin, chief data officer at Suffolk, was also a valuable resource and someone who could help with structuring data to properly to accurately predict outcomes.
"What we absolutely can show is that there's demonstrable evidence that suggests that some of the things that we're tracking are absolutely predictors of incidents," said Tim Stroud, national chief operating officer at Suffolk. "We use Newmetrix to capture information about what's going on on our project sites and we're able to think about and test our hypotheses around what are the things that are driving us to have safety incidences on a job site?"
Stroud said Suffolk can now accurately identify situations such as where project managers might have a higher-risk structural concrete activity alongside of a situation where they may have a diluted workforce due to seasonal holiday time off and its project managers can then act accordingly to mitigate that risk.
With Newmetrix becoming part of a much broader platform and scale with Oracle, Kanner, who previously co-founded Vela Systems before it was acquired by Autodesk in 2012, said he believes those AI-based insights can further help contractors of all sizes.Stroud said that Suffolk had supported the company as more than a customer over the years and was glad that it was being acquired by a larger player in Oracle and that it would be able to scale further across construction.
"We were actually a little bit worried that nobody might pick up Newmetrix," Stroud said. "They are doing something that is unique, which is taking the concept of predictability and being able to aggregate an industry's worth of data at some point in time in the future to really have a much better data set and, and predictive ability."
Both Kanner and Webster said they plan to use Newmetrix's predictive analytics capabilities across the Oracle Construction Intelligence Cloud.
"Product and team are coming in as a part of the broader construction intelligence platform," Webster said.
The platform, today, has descriptive as well as predictive analytics capabilities. Predictive behavior for schedule risk will immediately be complimented by adding Newmetrix's safety risk assessments, Webster said.
"The Newmetrix product is deep and has shown its value in the safety arena. We're really excited to see where it can go from here, as the team and the product and our expertise around how analytics is applied more broadly," he explained.
Newmetrix was one of the first AI platforms devoted to the safety component of construction risk. With a market that now includes predictive AI analysis products such as Doxel, Disperse.io and nPlan, Kanner said that the resources of Oracle will allow it to create better analyses of all aspects of construction.
"The construction industry is such a big industry. There's so much opportunity. we don't see it as a zero-sum game, if the goal really is helping [the] industry move forward," Kanner said. "I'm really enthusiastic about all of the innovation that's happening in construction as a whole."