Mixing AI With Concrete and Cement: A Q&A With Eric Koehler of Titan America

With concrete and cement technology going back thousands of years, seen in such famous historical structures as Rome’s Colosseum and Pantheon, you might think that builders have concrete pretty well figured out. Eric Koehler, vice president of innovation and product quality for Titan America, a building materials manufacturer operating across the Eastern United States, is here to tell you otherwise.

Koehler, a Fellow of the American Concrete Institute, is a named inventor on 16 concrete-related patents and was named one of the “Most Influential People in Concrete Production” by Concrete Producer magazine in 2016. He also holds a Ph.D. in civil engineering from the University of Texas at Austin. Prior to joining Titan America, Koehler worked in research and development for concrete quality management, focusing on chemical admixtures, sensors and automation.

We had the opportunity to sit down with Koehler to discuss innovations in cement and concrete and how these advancements intersect with AI to create smarter designs and smart buildings.

Do you have any project highlights or favorite things that you’ve been able to work on?

The first would be development of new cements that are both high performing and have lower emissions. (Titan) recently got a grant from the Department of Energy for over $60 million to build a new, first-of-its-kind cement manufacturing line in one of our plants in Virginia. So, I’ve spent a lot of my time working on that grant and then the engineering (of that cement manufacturing line). We’re still in the engineering phase right now, but we will start construction next year and then go through the commissioning.

A lot of the work has been on lowering our emissions, and we do that through a variety of ways. Now that we have a new president, we don’t talk as much about low CO₂—it’s almost a dirty word that you can’t say. But, I think for us, it’s about expanding the availability of cement and concrete, which are critical to the construction industry. It’s also about innovating the ways we make cement and making sure that we can make the cement domestically in the US.

Another area where we’ve done a lot of work is with AI and digitization of our operations and the way we interact with our customers. I think both of those topics are highly relevant when we talk about (the Buildings Conference) and the topics that we are focusing on at this particular conference.

How did you get started working on smart buildings? And do you feel like the industry has a standard definition for a smart building?

Smart buildings is such a broad term that you almost have to update the definition every year because what was smart last year is now just a standard thing. (Titan) got started with smart buildings more from a manufacturing perspective, so we don’t think about it in terms of electronics in the building but more as getting the building done faster using materials that perform better than commercial materials.

We operate two cement plants in the U.S. that are each billion-dollar manufacturing investments, as well as a series of ready-mix concrete plants from the Mid-Atlantic down to Miami, Fla. In order to get the most effective utilization out of those plants, we’ve been able to apply digitalization there. We’ve also started thinking about how we apply digitization further downstream with our customers who are contractors and developers. We want to use digital solutions to enhance the productivity of the contractor by making sure the concrete arrives on time and has higher performance attributes.

You mentioned using AI to determine curing time. Can you say more about where AI is intersecting with cement or concrete construction?

Internally, we do a lot of predictive maintenance process optimizers within our plants that are constantly taking in data from thousands of sensors and using it to actually run the plant. So the plant runs entirely on AI.

There are companies that make embedded sensors that monitor the concrete as it’s curing in the building, and we work with a number of those. Essentially, we can tie our AI tools into those sensors and be able to better predict the performance, which is one way we utilize AI externally, as well.

One example is a larger data center customer that came to us with goals for CO₂ reduction for one center in particular. We used AI to optimize the mix for them and were able to give them a much lower CO₂ with the same performance. We were also able to predict all the different properties of the concrete we were delivering. Then, we provided a dashboard that showed the building’s CO₂ footprint in real time so they didn’t have to wait till the end of the project to see the amount of CO₂ reduced. Now, they get an emailed report from us every Monday morning at 8 a.m. of exactly how much CO₂ they saved over the past week.

We also use AI for automatic scheduling optimization for delivery to our customers. There are a lot of factors that impact the schedule in real time. Customers may over-order what they need, orders might be canceled, there could be bad weather and other things. We only have a set number of trucks that we try to utilize efficiently. So, AI does the dynamic planning and routing of the trucks.

When we develop these tools, we think about how they’ll be used in our own operations or to benefit our own customers, but (Titan) also has a startup division called CemAI that is commercializing tools developed internally. I think this is an interesting model for the construction industry, because here’s a scenario where we can build technologies and then license and sell the software in other parts of the world where we don’t operate. We’ve sold CemAI in Africa, Latin America, and even on the West Coast of the United States. Companies who operate in only a couple of states or just a region aren’t competing with the rest of the country. It’s a model that’s really applicable to a fragmented market like construction.

With the new U.S. administration, are you seeing any changes in what customers are asking for regarding CO₂?

There’s still some demand for that, but I think there’s definitely people who have pulled back and said, “We’re going to wait, and we’re not going to emphasize this.” I think it’s too early to tell for sure.

A lot of concrete is utilized in transportation infrastructure, so there were lots of grants out there for low-carbon materials that have already been canceled by the new administration. Since the transportation infrastructure sector is the biggest user (of concrete), it affects the rest of the market, as well.

Clearly, there’s still interest out there and there’s still customers asking for this. So, I think a lot of it will go forward, but some of the subsidies and things from the government are not going to be there.

In talking about how the definition of smart buildings changes year to year, what is exciting or intriguing about this year’s definition?

The ability to integrate AI allows us to do everything more efficiently. From our standpoint, we can deploy new things faster. It doesn’t take as long to code and develop these technologies and so we can deploy digital tools for our customers faster and do more. I think people don’t care that it’s a smart building or a dumb building, or whatever the definition is, if we are able to show value to the customer through their savings or some other benefit. So, we can get these things faster and show a benefit over the current state of the art: they can open their building faster and then have less maintenance in the building.

Another big thing is that the number of different materials out there in the world has changed.

Speaking of materials, what changes have you seen in concrete, cement, fly ash and aggregates in the last five to 15 years that have really made a big impact in the industry?

I think the biggest thing has been the sheer amount of materials that are out there now. We went from an industry that had essentially one type of cement to now having all these new types of materials.

If you look at just the growth of the industry, we need more of these innovative materials to help meet the demand for construction. There’s been a lot of startups that have come into the space. As I mentioned before, (Titan) has a corporate venture capital arm that’s invested in several of these companies. I think building owners, engineers, and designers are often perplexed by the sheer number of new materials that are out there and how to incorporate them. People will read an article about some kind of material that they heard about that someone made just a couple grams of in the lab at some university. We add a lot of value because of our ability to take these new materials and integrate them into mixes for concrete that performs well and meets all the other criteria of the project.

Would you consider the development of different building materials a tipping point that propelled smart buildings forward?

Yes. People have different reasons for why they wanted to integrate these new materials and I think there are three big areas for why. Some wanted lower carbon, so a lot of these materials have that as a characteristic. Some just wanted a faster construction time so builders can move on to the next stage of construction faster; so, some of these materials enable that. Others want a material where there’s potentially less maintenance costs in the building over time, and for that we need to be able to quantify that and have a material that is going to work over the long term to reduce maintenance costs.

What future advancements are you seeing that you’re excited for?

I think a lot of the future work that we’re very excited for centers around circularity and the circular economy. We already do a lot of that today, but we’d like to do more. For example, the main obvious instance is construction and demolition waste. As older buildings are torn down, the industry is challenged to find ways to integrate the waste into new buildings and our processes.

Titan already reuses a lot of solid waste and other kinds of material as fuels or actual raw materials in our product, and we’d like to use even more of that waste. There’s actually a lot of startups out there that are making that possible and devising new ways to utilize old materials. We’re very interested in that concept and working with some of them.

What are some common misconceptions that people have around concrete and cement? And are any of these related to smart buildings?

Smart buildings or not, people seem to think (concrete and cement are) all the same: gray. But, there’s more and more materials out there that can enable us to do more and more things to meet the criteria. I think one of the areas that we always struggle with is getting in the design process early enough so that people know what’s possible and are able to choose the best material with enough time to create a whole design around it. If people come to us as the material supplier two weeks before they need to start construction, there’s not a lot of options compared to if they start earlier.