Giatec announced on Tuesday that their artificial intelligence program, Roxi, has been trained with the ability to detect possible errors in concrete maturity calibrations and mix proportions. Roxi was the initial AI program created for concrete testing. It was developed in conjunction with Montreal Institute of Learning Algorithms, Mila, Quebec’s Artificial Intelligence institute.
With the help of Mila and Giatec’s team of AI experts, Roxi provides new insights into concrete curing by using machine learning algorithms. With this latest update, Roxi can uncover human error that may occur when concrete mixes are set up for maturity calibration, offering suggestions to correct calibration data. This includes the ability to analyze concrete mix proportions such as specified strength, slump, and air content.
Giatec Scientific is strengthening the construction industry by bringing smart testing technologies and real-time data collection to the forefront of every jobsite. Giatec’s suite of hardware and software products has leveraged advanced technologies such as artificial intelligence (AI), and Internet of Things (IoT), including wireless concrete sensors, mobile apps, and advanced non-destructive technologies (NDT) to drive innovation throughout concrete’s lifecycle.
Giatec wants to acknowledge the generous support it received from the National Research Council of Canada Industrial Research Assistance Program (NRC IRAP) and Sustainable Development Technology Canada (SDTC) in the development of the Roxi AI program.
“My team and I are excited to be working with Giatec, a company that has a revolutionary approach to leveraging artificial intelligence and machine learning to solve real-world problems in the construction industry,” says Frédéric Laurin, director of partnerships at Mila.
Roxi was built for Giatec’s SmartRock wireless concrete sensor. Through these sensors, currently being used in over 6,000 construction projects worldwide, Roxi has access to millions of data points. As the first truly wireless sensor introduced to the market in 2015, SmartRock is the leading concrete maturity sensor for accurate monitoring of concrete curing and hardening.
Unlike time-consuming and error-prone break tests, or cumbersome wired sensors, Giatec’s patented maturity sensor uses a highly accurate ASTM-approved testing method. This has translated into faster, safer, and more economical concrete construction.
Training Roxi for mix calibration verification was done with the help of mix data provided by Stoneway Concrete, Central Concrete (a U.S. Concrete company), Maschmeyer Concrete Company of Florida, Hilltop Concrete and Tomlinson.
“We are very happy to have collaborated with Giatec in the development of the first-ever concrete testing AI assistant,” said Patrick Frawley, director of quality assurance at Central Concrete. “Now, we can assure our contractors that their data is verified because our team can easily and confidently calibrate our concrete mixes for maturity.”