Posted on December 29, 2022
In the 20th century, the question “Can machines think?” captured the imagination of some of the planet’s greatest thinkers. One was Alan Turing, the young British polymath who dared to answer “Yes!” His 1950 paper, “Computing Machinery and Intelligence,” suggested that machines, like humans, could learn to reason and solve problems. In certain ways, what he theorized seven decades ago is now a reality. Artificial intelligence (AI) impacts every major industry, and concrete is no exception.
Today, concrete professionals increasingly look to AI for:
Enhanced concrete production
Better concrete performance
More effective concrete testing
Major associations also focus their attention on AI. For example, “The Concrete Industry in the Era of Artificial Intelligence” was a Technical Session held at the 2020 Virtual Concrete Convention of the American Concrete Institute (ACI).
Together, population growth and climate change demand more resilient buildings and infrastructure. The concrete industry is uniquely positioned to heed the call. At the same time, it is vital to move every segment of the industry toward carbon neutrality. The emergence of wireless sensors and millions of data points advances this quest. Now, artificial intelligence and machine learning make even better use of all the available data.
How far can AI take an industry seeking to meet 1) growing global demand, and 2) the demands for carbon neutrality?
AI and Concrete Production
Concrete.ai offers a SaaS solution that optimizes ready-mix and precast concrete production. In September 2022, Concrete.ai released the beta version of its new AI software platform. The tool “delivers unparalleled reductions” in the embodied carbon intensity of concrete projects. It cuts costs by “utilizing locally available raw materials to ensure safety, longevity, and code-compliance.”
Pilot tests delivered verifiable embodied carbon reductions of 12% to 70%. Embodied carbon accounts for all CO2 emissions throughout the production cycle. This includes extraction, manufacturing, transportation, construction, and end-of-life processing. Its ultimate goal is to reduce CO2 emissions by up to 500 million tons per year.
AI and Concrete Mixing
Properly deployed, AI improves efficiencies in cement production while ensuring quality. This helps to overcome a variety of challenges in concrete production.
For example, traditional concrete mixing must overcome a series of possible outcomes. Some examples include:
Honeycombing – exposed coarse aggregate
Bugholes – too lean a mix leads to small surface holes
Scaling – due to improper proportions of concrete’s constituent parts
These problems can increase water infiltration due to the freeze/thaw cycle.
Case Study: Marcotte Batch
Sophisticated software has already improved concrete mixing for more than two decades. Marcotte Batch is one example. It is in use at more than 1100 plants worldwide.
With the help of AI, Marcotte Batch for Ready Mix adjusts the amount of each batch component in real time. It accomplishes this while maintaining key ratios and volumetric yields. AI also optimizes production processes and predicts failure. It enhances predictive maintenance as well.
As Marcotte notes, “AI research is blossoming in concrete science and engineering, where it has offered new insights towards mixture design optimization and service life prediction of cementitious systems.“
Today's concrete professionals use AI to deliver further efficiencies in concrete mixing.
Future of AI in the Concrete Industry
Giatec uses AI to make its concrete testing platform Roxi even more intuitive. AI will also offer more insight into how concrete will perform further into its life cycle.
Giatec says its technology only needs two days to predict concrete strength. This means project managers will have essential information ahead of schedule.
Retooling existing infrastructure, while also vital, is expensive and time-consuming. By comparison, software-based solutions are readily deployed around the world.
AI and Concrete Testing
Researchers examine AI’s contribution to various kinds of concrete testing. For example, they use AI models to estimate the compressive strength of concrete. One study examined AI's ability to determine the compressive strength of concrete with fly ash. Another study looked as concrete with silica fume as an SCM. AI has also been used to analyze the compressive, tensile and flexural strengths of concrete with metakaolin.
The power of machine learning has also been harnessed to improve the self-healing capacity of concrete. AI also made it possible to optimize the use of the elastoacoustical effect for stress monitoring.
Case Study: Roxi
In 2022, Giatec introduced Roxi, the first AI program for concrete testing. Roxi is designed to speed up the construction process. To develop Roxi, Giatec collaborated with Mila (Montreal Institute of Learning Algorithms). Milk is a community of more than 1000 machine learning specialists in Quebec, Canada.
Roxi gets its data from wireless sensors secured to rebar within concrete formwork. These sensors wirelessly transmit temperature data to an app on smart devices. A maturity equation in the app allows contractors to calculate compressive strength. Roxi yields accurate strength data in real time.
Contractors use the Giatec 360 Dashboard to confirm the consistency of curing in the core and on the surface of a slab. They know more precisely when to remove formwork, or when to perform post-tensioning. Roxi also exposes anomalies at various points in the concrete life cycle.
The Pennsylvania Aggregate and Concrete Association (PACA) reports on industry innovation through SpecifyConcrete.org. Our team welcomes the opportunity to answer questions about your upcoming concrete project. Please contact us at your convenience.